Thermograph network in the Gulf of St. Lawrence

B. Pettigrew, D. Gilbert and R. Desmarais

Sciences Branch Fisheries and Oceans Canada Maurice Lamontagne Institute 850 route de la Mer, P.O. Box 1000 Mont-Joli (Québec) Canada, G5H 3Z4

2016

Canadian Technical Report of Hydrography and Ocean Sciences 311

Canadian Technical Report of Hydrography and Ocean Sciences

Technical reports contain scientific and technical information of a type that represents a contribution

to existing knowledge but which is not normally found in the primary literature. The subject matter is

generally related to programs and interests of the Oceans and Science sectors of Fisheries and Oceans

Canada.

Technical reports may be cited as full publications. The correct citation appears above the abstract

of each report. Each report is abstracted in the data base Aquatic Sciences and Fisheries Abstracts.

Technical reports are produced regionally but are numbered nationally. Requests for individual

reports will be filled by the issuing establishment listed on the front cover and title page.

Regional and headquarters establishments of Ocean Science and Surveys ceased publication of their

various report series as of December 1981. A complete listing of these publications and the last number

issued under each title are published in the Canadian Journal of Fisheries and Aquatic Sciences, Volume

38: Index to Publications 1981. The current series began with Report Number 1 in January 1982.

Rapport technique canadien sur l'hydrographie et les sciences océaniques

Les rapports techniques contiennent des renseignements scientifiques et techniques qui constituent

une contribution aux connaissances actuelles mais que l'on ne trouve pas normalement dans les revues

scientifiques. Le sujet est généralement rattaché aux programmes et intérêts des secteurs des Océans et

des Sciences de Pêches et Océans Canada.

Les rapports techniques peuvent être cités comme des publications à part entière. Le titre exact

figure au-dessus du résumé de chaque rapport. Les rapports techniques sont résumés dans la base de

données Résumés des sciences aquatiques et halieutiques.

Les rapports techniques sont produits à l'échelon régional, mais numérotés à l'échelon national. Les

demandes de rapports seront satisfaites par l'établissement auteur dont le nom figure sur la couverture et

la page de titre.

Les établissements de l’ancien secteur des Sciences et Levés océaniques dans les régions et à l'ad-

ministration centrale ont cessé de publier leurs diverses séries de rapports en décembre 1981. Vous

trouverez dans l'index des publications du volume 38 du Journal canadien des sciences halieutiques et

aquatiques, la liste de ces publications ainsi que le dernier numéro paru dans chaque catégorie. La

nouvelle série a commencé avec la publication du rapport numéro 1 en janvier 1982.

Canadian Technical Report of

Hydrography and Ocean Sciences 311

2016

THERMOGRAPH NETWORK IN THE

GULF OF ST. LAWRENCE

by

B. Pettigrew, D. Gilbert and R. Desmarais

Sciences Branch

Fisheries and Oceans Canada

Maurice Lamontagne Institute

850 route de la Mer, P.O. Box 1000

Mont-Joli (Québec)

Canada, G5H 3Z4

ii

© Her Majesty the Queen in Right of Canada, 2016.

Cat. No. Fs 97-18/311E-PDF ISBN 978-0-660-05537-4 ISSN 1488-5417

Correct citation for this publication :

Pettigrew, B., Gilbert, D. and Desmarais R. 2016. Thermograph network in the Gulf of St.

Lawrence. Can. Tech. Rep. Hydrogr. Ocean Sci. 311: vi + 77 p.

iii

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................................... iv

LIST OF FIGURES .........................................................................................................................v

ABSTRACT ................................................................................................................................... vi

RÉSUMÉ ....................................................................................................................................... vi

1. INTRODUCTION .......................................................................................................................1

2. METHODS ..................................................................................................................................3

2.1 LOGISTICS ............................................................................................................................3

2.2 INSTRUMENTS AND CALIBRATION ...............................................................................5

2.3 DATA PROCESSING AND VALIDATION ........................................................................6

2.4 INSTRUMENT DEPTH ADJUSTMENT AND OTHER RELEVANT DETAILS ............13

2.5 CHANGE IN STATION POSITION ...................................................................................18

3. RESULTS ..................................................................................................................................19

3.1 MISSING DATA ESTIMATION.........................................................................................19

3.2 YEARLY DATA AVAILABILITY CHART ......................................................................22

3.3 SPRING–NEAP TIDAL CYCLE AT BIC STATION .........................................................22

3.4 WIND EFFECT AT LA TABATIÈRE STATION ..............................................................23

4. DISCUSSION ............................................................................................................................25

4.1 THE OVERSAMPLING ISSUE ..........................................................................................25

4.2 THE MISSING RECORD MODEL .....................................................................................26

5. CONCLUSIONS........................................................................................................................26

6. ACKNOWLEDGEMENTS .......................................................................................................27

7. REFERENCES ..........................................................................................................................27

APPENDIX 1. THERMOGRAPH NETWORK PHOTO LIBRARY ..........................................29

APPENDIX 2. TABLES OF RECORD NUMBERS ....................................................................35

APPENDIX 3. CHARTS OF DATA AVAILABILITY ...............................................................57

iv

LIST OF TABLES

Table 1. Thermograph network stations. In the “Ending year” column, ---- means still

in operation. H is the water depth, and the nominal instrument depth is given

by Z. The last column indicates the platform instrument owner (PIO) where

CCG, EC, MLI, and PB identify respectively maritime traffic CCG buoy,

meteorological buoy of Environment Canada, MLI buoy or mooring, and

private buoy. The Belle Isle station has an ADCP close to the sea bottom that

measures currents throughout the water column; bottom water temperature as

well as salinity and pressure are measured year-round. The three stations with

an * are opportunistic and may be removed at any time. The buoy at

Natashquan was moved during the first year of operation. Station equipped

with a MOB is in bold type .........................................................................................4

Table 2. Details of temperature sensor models used in the MTN. The column labelled

“Period of Use” gives the range of years over which each model was used. N

is the number of times series recorded with the specified instrument during the

1993–2014 period. There is a grand total of 964 seawater temperature time

series recorded during that period. Resolution and accuracy were provided by

the manufacturers in their instrument specifications except for the Vemco

Minilog12 whose resolution and accuracy where determined at MLI .......................7

Table 3a. Some instantaneous seawater temperatures recorded by the main and backup

instruments at La Perle station on 5 September 2007 at 26 m ..................................11

Table 3b. Some instantaneous seawater temperatures recorded by the main and backup

instruments at La Perle station on 9 August 2007 at 26 m. ......................................11

Table 4. Height (L) and weight of CCG buoy anchors used at thermograph network

stations. Station name is followed by sea bottom type in parenthesis, where

MB, RB, and SB mean respectively muddy bottom, rocky bottom and sandy

bottom. The Grande-Rivière station was equipped with a CCG buoy until

2011...........................................................................................................................13

Table 5. Water depth corrections. The station name is followed by the former depth

and the corrected depth values. Corrections in the raw file database were

carried out on 8–9 July 2009 and 7–8 January 2016 ................................................14

Table 6. Positions of the EC meteorological buoy at Mont-Louis station from 1994 to

2014...........................................................................................................................17

Table 7. Quality control flag meanings ...................................................................................20

Table 8. Record number definitions ........................................................................................21

v

LIST OF FIGURES

Figure 1. Thermograph network station positions during the 1993–2014 period. The

inset map shows the northernmost year-round stations, Hudson-1 and

Hudson-3, as well as the seasonal Churchill station on the west coast of

Hudson Bay. Year-round station names are in bold type. The 200 m isobath

is plotted .........................................................................................................................2

Figure 2. Distribution of seawater temperature differences for the Minilog12 for pairs

of instruments deployed at the same depth, location and time period. Only

the values in the interval ±0.25°C are shown. There are 415 080 values with

a mean difference of 0.00074°C, a standard deviation of 0.03°C, and a

median of 0.001°C .........................................................................................................9

Figure 3. Seawater temperature differences for the Minilog12 outside the ±0.25°C,

interval. There are 2484 values in this interval, with a minimum of -5.417°C

and a maximum of 1.907°C. The mean value difference is 0.034°C, the

standard deviation is 0.463°C, and the median is -0.264°C ........................................10

Figure 4. Seawater temperature differences for the Minilog16. There are 111 072

values, that come from 7 stations and 12 pairs of instrument, recorded from

2011 to 2013. The mean value difference is 0.002°C, the standard deviation

is 0.043°C, and the median is 0.001°C ........................................................................12

Figure 5. Instrument depth at Bonne Bay station ........................................................................17

Figure 6. Seawater temperature differences between the original Grande-Rivière site

and the proposed Grande-Rivière site at 2 m depth .....................................................18

Figure 7. Temperature at Bic station at 1 m depth in 2007. The blue curve shows all

data while the red curve is using a 48-hours triangular filter. The black full

circles indicate the new moon while the empty circles indicate the full

moon. The vertical dotted lines pinpoint the local non-filtered SST

minimum for each spring-neap tidal cycle. The lower black curve is the

hourly water level (divided by 2 to improve graphic visualization) measured

at Rimouski by a Canadian Hydrographic Service tide gauge (station no.

2985) ............................................................................................................................23

Figure 8. Seawater temperature at La Tabatière station in 2014 at 1 m (blue dotted

line) and 39 m (black line). The black full circles indicate the new moon

while the empty circles indicate the full moon (top). Daily wind measured at

Blanc-Sablon in 2014 by AES. The wind direction is defined here as the

direction toward which the wind blows. Only winds higher than 32 km/h are

shown (bottom) ............................................................................................................24

vi

ABSTRACT

Pettigrew, B., Gilbert, D. and Desmarais R. 2016. Thermograph network in the Gulf of St.

Lawrence. Can. Tech. Rep. Hydrogr. Ocean Sci. 311: vi + 77 p.

The thermograph network program of the Maurice Lamontagne Institute started in 1993 with six

monitoring stations. The main aim was to measure seawater surface temperature in the Estuary

and Gulf of St. Lawrence and to maintain a database of those data. Today’s measurements are no

longer limited to surface water temperature: some stations are now equipped with multi-devices

platforms allowing the measurement of oceanographic, optical and meteorological parameters.

However, we focus our attention in this report exclusively on the seawater temperature records

with the goal of facilitating their use by identifying available data as well as missing data

records. We have tabulated relevant record numbers that characterize the completeness of the

raw data files recorded during the period from 1993 to 2014. To complement these tables we

plotted figures for every year that show the time intervals with valid data records for each

station. We also present some examples of physical oceanographic phenomena in the coastal

waters of the program area.

RÉSUMÉ

Pettigrew, B., Gilbert, D. and Desmarais R. 2016. Thermograph network in the Gulf of St.

Lawrence. Can. Tech. Rep. Hydrogr. Ocean Sci. 311: vi + 77 p.

Le programme du réseau de thermographes de l’Institut Maurice-Lamontagne a débuté en 1993

avec six stations de monitorage. Le principal objectif consistait à mesurer les températures de

surface de la mer dans l’estuaire et le golfe du Saint-Laurent tout en maintenant une base de

données. Aujourd’hui les mesures ne sont plus limitées aux températures de surface de l’eau;

certaines stations sont munies de plates-formes multi-instruments permettant la mesure de

paramètres océanographiques, optiques et météorologiques. Cependant dans ce rapport nous

mettons l’emphase exclusivement sur la disponibilité des enregistrements de température de

l’eau dans le but d’en faciliter leur utilisation en identifiant les données disponibles de même que

les données manquantes. Nous avons écrit dans des tableaux certains des nombres pertinents

d’enregistrements qui caractérisent le degré d’intégralité des fichiers bruts enregistrés durant la

période de 1993 à 2014. Pour compléter ces tableaux nous avons tracé pour chaque année une

figure montrant les enregistrements valides pour chaque station. Nous présentons aussi quelques

exemples de processus d’océanographie physique dans les eaux côtières du programme.

1

1.0 INTRODUCTION

The Global Climate Observing System (GCOS-164, 2012) describes sea surface temperature

(SST) as an essential climate variable because of the key role that SST plays in air-sea

interactions. Long-term SST time series require time (typically, more than 30 years) to attain a

certain level of maturity but such time series are still relatively rare. Meanwhile, computer

advancements have greatly facilitated the management and use of large amounts of data.

However, there is no guarantee that any time series that started twenty years ago or so will

survive the numerous obstacles that it will face, such as costs related to mooring installation and

recovery, instrument purchase, instrument calibration and maintenance, data processing, and

changes in corporate priorities.

In 1978, the Bedford Institute of Oceanography (Fisheries and Oceans Canada [DFO],

Dartmouth NS) started measuring water temperature in some Gulf of St. Lawrence locations. At

that time, they were using electro-mechanical thermographs (Ryan model; see Figure A1.1)

capable of recording temperature for a period of a few months with an accuracy of 0.5°C. Ten

years later, the Maurice Lamontagne Institute (MLI), a new DFO oceanographic research centre,

took charge of monitoring surface water temperature in the Estuary and Gulf of St. Lawrence

(hereafter EGSL). From 1988 to 1992, thermographs were installed by fishermen close to their

fishing gear. This way of measuring water temperature cannot be considered a systematic

monitoring program because fishermen did not necessarily place their fishing gear at the same

location every year. Furthermore, fishing gears were sometimes moved even during the same

year to increase catches. Coastal areas surrounding the EGSL have high SST variability caused

by tides, winds, freshwater runoff, bathymetry, and shallow water. Such high SST variability

over relatively short horizontal distances emphasizes the importance of fixed stations. The data

recorded from 1988–1992 should be used carefully because of large uncertainties for depth and

position. We do not report on those data here.

For all the above reasons, it was decided in 1993 to set up fixed-sites monitoring program

resulting in what is known as the MLI thermograph network (MTN). The initial goal of this

program was to create a database of coastal SST in the EGSL. We started the program with six

sampling stations and gradually increased the number of stations to obtain a better spatial

coverage of the study area (Figure 1).

This monitoring program is greatly indebted to the Canadian Coast Guard (CCG). Indeed, over

the years we have used maritime traffic buoys as platforms for installing most of our

thermographs. Because of mooring costs it would have been impossible to proceed without this

partnership with the CCG. Since 1996, we have used robust steel supports for our thermographs

that were especially built to be installed close to the CCG buoy anchor. All CCG buoy stations

currently measure SST and sea bottom temperature (SBT) along with temperature at intermediate

depths at some stations. To fulfill some specific scientific needs including the validation of

satellite-derived SST and ocean color, MLI technicians have built oceanographic buoys (MOB:

MLI Oceanographic Buoy) to be moored at locations that are further offshore than the usual

CCG maritime traffic buoys. These MOBs are equipped with oceanographic, meteorological and

optical sensors. Here again, we benefited from our partnership with the CCG, who installed and

2

recovered these buoys during their normal operations in the spring and fall. We also maintain a

year-round mooring in the Strait of Belle Isle that measures ocean currents with an Acoustic

Doppler Current Profiler (ADCP) installed on the seafloor; bottom temperature, salinity and

pressure are also recorded.

Figure 1. Thermograph network station positions during the 1993–2014 period. The inset map

shows the northernmost year-round stations, Hudson-1 and Hudson-3, as well as the seasonal

Churchill station on the west coast of Hudson Bay. Year-round station names are in bold type.

The 200 m isobath is plotted.

An increasing need for SST time series of at least 30 years in duration mainly comes from the

numerous research projects on climate. None of our time series has that minimum age yet, but a

few are getting close. Nevertheless some SST time series, like the ones recorded by the

oceanographic buoys, are very useful for validating the Advanced Very High Resolution

Radiometer (AVHRR) SST products processed at MLI (Pettigrew et al., 2011). Water

temperature can also be very useful for biologists in their studies on commercial species (e.g.

lobster, crab, scallop). In situ SST data can also contribute to the validation of hydrodynamic

models (Saucier et al., 2003). It is worth mentioning that a MOB has real-time transmitting

capability, so it is possible for these data to be used in environmental emergencies or during a

CCG Search and Rescue event.

3

Statistics from MTN seawater temperature time series, e.g., monthly means were described for

the first time in Gilbert et al. (1997). They are also regularly included in research documents on

oceanographic conditions in the EGSL (e.g., Galbraith et al., 2015).

The main goal of this report is to help MTN data users by documenting the presence of gaps. The

visualization of gaps will rapidly inform the reader about time-series completeness, the number

of gaps and their maximum length. Finally some oceanographic processes present in the EGSL

will be showcased using seawater temperature time series from the network.

2.0 METHODS

2.1 THE LOGISTICS

In this section, we explain the various logistical steps that take place between programming an

instrument, its installation on a CCG buoy, and the publication of the resulting seawater

temperature time series. The field logistics of the thermograph network (Figure 1, Table 1) is an

important component of this program, and it involves a close collaboration between MLI

employees and staff of the CCG Aids to Navigation Program of the Québec and Maritimes

regions, as well as CCG ship crew members. Aids to Navigation provide visual aids such as

lights, beacons, and surface buoys.

We have equipped 11 maritime traffic buoys with our water temperature recorders. A robust

steel thermograph support is welded onto the buoy, allowing water temperature to be measured

between 0.5 and 2.0 m of depth depending on the shape or type of the CCG buoy. To measure

bottom temperature, we use a shackle to attach a double thermograph support (Figure A1.2)

directly on the sinker. We initially thought that the use of a support for two instruments was

justified because recovering a bottom instrument is likely more risky and damage is sometimes

inevitable, so the double support would maximize our chances of getting data. But, in fact, the

thermographs attached to the surface buoy are equally at risk: the thermograph support welded

onto the buoy side can be damaged when the buoy is stowed in the ship’s hold before installation

or after recovery. Although initially we thought that using CCG buoys was risky for both surface

and bottom instruments, we found after more than 20 years that the risk turned out to be minor.

We will see later that the dual bottom instruments can be used to check instruments accuracy.

Thermographs are installed on buoys by MLI staff before the buoys are taken on board whereas

the thermographs close to the anchor are installed by CCG ship crew just before buoy

deployment. The instruments supplied to CCG by MLI staff are clearly identified to ensure that

each instrument will be installed on the right buoy.

We targeted five sites (Banc Beaugé, Courant de Gaspé, Gyre Anticosti, Rimouski and Shediac

Valley) to be equipped with a MOB. These five sites constitute a separate network within the

thermograph network. A MOB has meteorological, oceanographic and optical sensors; they can

transmit and receive data using a cellular link that has a working range estimated at 15 km from a

cell antenna, a UHF modem if the buoy is less than 30 km from a terrestrial reception antenna, or

a satellite link that has no distance limitation. Shediac Valley is a high frequency Atlantic Zone

4

Table 1. Thermograph network stations. In the “Ending year” column, ---- means still in

operation. H is the water depth, and the nominal instrument depth is given by Z. The last column

indicates the platform instrument owner (PIO) where CCG, EC, MLI, and PB identify

respectively maritime traffic CCG buoy, meteorological buoy of Environment Canada, MLI

buoy or mooring, and private buoy. The Belle Isle station has an ADCP close to the sea bottom

that measures currents throughout the water column; bottom water temperature as well as salinity

and pressure are measured year-round. The three stations with an * are opportunistic and may be

removed at any time. The buoy at Natashquan was moved during the first year of operation.

Stations equipped with a MOB are in bold type.

Station Starting

year

Ending

year

Latitude

N

Longitude

W

H

m

Z

m

PIO

Baie-Comeau 1999 ---- 49° 12.07 68° 03.35 82.3 1, 82.3 CCG

Banc Beaugé 1998 ---- 49° 30.73 60° 04.32 97 0.5, 1, 97 MLI

Belle Isle 2007 ---- 51° 34.80 56° 37.20 75 71 MLI

Bic 1994 ---- 48° 19.95 69° 09.92 30.5 1, 2, 30.5 CCG

Blanc-Sablon 1999 ---- 51° 23.97 57° 11.65 22.0 1, 22 CCG

Bonne Bay 1993 2004 49° 32.30 57° 56.00 25.0 0.5, 25 MLI

Borden 1993 ---- 46° 14.96 63° 42.06 1.5 1.25 MLI

Churchill 2006 2007 58° 49.67 94° 06.27 30 0.5 PB

Courant de Gaspé 2005 ---- 49° 14.50 66° 12.00 165 0.5, 165 MLI

Grande-Rivière 1994 ---- 48° 23.33 64° 29.53 10 0.5, 2, 10 MLI

Gyre Anticosti 2005 2011 49° 43.00 66° 15.00 337 0.5, 337 MLI

Havre-St-Pierre 1996 ---- 50° 06.50 63° 38.50 120 1, 2, 120 MLI

Hudson-1* 2013 ---- 60° 38.53 65° 52.08 270.7 268.4 MLI

Hudson-3* 2012 ---- 62° 57.18 74° 41.13 161.7 159.4 MLI

IML 1994 ---- 48° 39.60 68° 09.39 14 0.5, 14 MLI

IML PEM 1994 ---- 48° 39.52 68° 09.39 13 12.4 MLI

Ile Shag 1993 ---- 47° 28.51 61° 41.30 10 1, 10 MLI

Irving Whale 1998 ---- 47° 24.20 63° 23.53 67 1, 67 MLI

La Perle 1996 ---- 47° 19.45 61° 34.30 26 1, 26 CCG

La Romaine 1994 ---- 50° 07.31 60° 18.58 21.9 1, 2, 21.9 CCG

La Tabatière 2002 ---- 50° 51.20 58° 57.01 39 1, 39 CCG

Mont-Louis 1994 ---- 49° 32.35 65° 42.64 325 0.35, 0.5, 1.1 EC

2.1, 30, 200

Natashquan 1994 ---- 50° 11.20 61° 50.67 5.4 1, 5.4 CCG

Natashquan 1993 1994 50° 11.00 61° 50.95 12 12 CCG

Old Belle Isle 2004 2006 51° 21.10 56° 52.50 111 105 MLI

Old Harry* 2014 ---- 48° 00.00 60° 30.00 450 0.5 MLI

Port-Menier 1994 ---- 49° 46.23 64° 20.62 12.8 2, 12.8 CCG

Rimouski 2002 ---- 48° 40.00 68° 35.00 330 0.5, 330 MLI

Riv.-au-Tonnerre 1998 ---- 50° 15.83 64° 46.79 16 1, 16 CCG

Sept-Iles 1993 ---- 50° 10.29 66° 25.76 21.9 1, 2, 21.9 CCG

Shediac Valley 2004 ---- 47° 47.00 64° 02.00 85.8 0.5, 85.8 MLI

Tadoussac 1993 ---- 48° 07.19 69° 40.50 36.6 1, 36.6 CCG

5

Monitoring Program (AZMP; see Therriault et al., 1998) station since 1998 and so is Rimouski

station, but since 2014 only. It is worth mentioning that Courant de Gaspé and Gyre Anticosti

used to be AZMP stations until 2013, after which they were replaced by the Rimouski station.

There is enough flash memory in the MOB controller to save all data recorded during a sampling

season. These buoys are the result of a partnership between MLI and a private company (Multi-

Électronique [MTE] Inc., Rimouski, QC, Canada); they were especially built to validate satellite

data like remote SST from AVHRR sensors and satellite water colour data from the Sea-viewing

Wide Field-of-view Sensor (SeaWiFS). The first generation MOB was installed at Rimouski

station in 2002. In spring 2013, the first new-generation “Viking” MOB was deployed at

Rimouski station; other stations should be similarly equipped in the years to come. In 2014, a

second Viking MOB was installed at Old Harry station. However, this is an opportunistic station,

meaning that the buoy can be retired at any time. We also installed instruments on the two other

opportunistic mooring stations, Hudson-1 and Hudson-3. In 2015, a third Viking buoy was

installed at Shédiac station to replace the first generation buoy. This Viking MOB has a bigger

hull than the first generation MOB, making the Viking a much more stable buoy that facilitates

secure at-sea maintenance. The MOB controller has been updated with more digital ports and

fewer analog ports. MLI staff deploy the MOBs from a buoy tender in spring and recover them

in late fall.

Additional sites were chosen to measure SST and bottom temperature to get better offshore

coverage. This explains the existence of the Havre Saint-Pierre and Irving Whale stations. CCG

decided in 2012 to eliminate their maritime traffic buoy at Grande-Rivière, which had been a

monitoring station since 1994. This site is now equipped with our own buoy to continue the

monitoring of seawater temperature at the same location. The station IML (measuring SST at 0.5

m depth and SBT at 13.7 m depth) is located 190 m north of the IML PEM station. The

instrument on the IML PEM station is installed by divers during the yearly fall maintenance of

the MLI seawater intake. Finally, we also have year-round sites at Shag Island and Borden,

where instruments are installed and recovered by local personnel. At the Shag Island station, an

instrument measuring SST during ice-free months is installed on a small buoy while SBT is

measured year-round. We measured SST (at 0.5 m) at Borden station from 1993 to 1995, then

from 1995 we started to measure SBT (at 1.5 m); however, monitoring was interrupted in 2003

and began again in spring 2008 to measure SBT (at 1.25 m). To confirm instrument depth at

Borden we used a Sea-Bird-37SM, which has a pressure sensor, in 2010. The lowest water

column height is about 1.2, whereas the highest water column height is around 3.9 m. We also

used to measure SST and SBT at a site in Bonne Bay (Newfoundland and Labrador).

Unfortunately the loss of our local field logistics person stopped that time series in 2004.

2.2 INSTRUMENTS AND CALIBRATION

The basis of data quality assurance is the careful monitoring of instrument calibration. For every

thermograph, a calibration file is produced that contains all calibrations done at various times,

allowing us to estimate sensor drift. Because instruments do not reproduce the same count

reading at different times of measurement for the same input signal, a calibration is done before

deployment and right after instrument recovery. These calibrations are very helpful when one

6

desires to determine the sensor drift that any device will experience over a period of time. A

common practice is to use the average of pre- and post-calibration to compute seawater

temperature. The comparison of temperatures obtained using both calibrations allows the

calculation of sensor drift.

Because sensor drift depends on temperature, we strive to choose two raw values: one at the

beginning of the time series and one that is 2–3 months later and for which seawater temperature

is often warmer than at the beginning of time series. Most of the time, instruments are very stable

and the drift is smaller than 0.001°C/month. Temperature sensor data that exhibit uncommon

drift (say > 0.05°C/month) are not included in our database. When this happens, we ask for a

calibration check; if the drift remains unchanged, the instrument is eliminated from the pool of

instruments and its data are not considered reliable.

The MLI calibration laboratory produced the coefficients used to transform raw values into

engineering values for all instruments deployed in this program except for instruments from Sea-

Bird Electronics. Because of the high accuracy (0.002°C) of these instruments, we can only

verify that the sensor has not drifted too much since the last verification. To do this, we compare

our temperature reference instrument (Rosemount manufacturer, Model 162 CE, serial number

3825) with the Sea-Bird sensor’s temperature value; the difference should be smaller than

0.01°C. If not, the instrument is sent back to the manufacturer for calibration. Sea-Bird

instruments are usually calibrated every 5 years.

Over the last 20 years, instrument resolution and accuracy have improved and memory has

increased, allowing us to store more and more data. Table 2 lists instruments (seen in Figure

A1.1) used in the MTN program, which has recorded 964 time series from 1993 to 2014. Certain

model resolutions are somewhat different from the specification published by the manufacturer;

this is the case when the temperature range for calibration is too wide for our purposes. If a

temperature sensor is calibrated over the range [-30°C, 40°C] by the manufacturer, then we will

reduce the range during calibration at MLI so that it will be more representative of water

temperatures found in our study area. A calibration temperature range like [-2.0°C, 22°C] instead

of [-30°C, 40°C] will have the effect of improving the original resolution and accuracy of the

instrument.

2.3 DATA PROCESSING AND VALIDATION

Some station metadata like location (latitude and longitude), instrument depth, date and time of

instrument installation and recovery, instrument type, and instrument serial number are written in

a logbook. This logbook is intensively used during data processing and any useful information

will be written down on the station log sheet.

To transform raw temperature retrieved from different instrument models into engineering units,

we used a homemade Graphical User Interface (GUI). This program writes the result in a

uniform file format called TS9, which has a 59-line header of metadata followed by data

columns of record time, water temperature, water temperature quality control (qc) flag, pressure,

pressure qc flag, conductivity, conductivity qc flag, depth, depth qc flag, salinity, salinity qc flag,

7

density anomaly (σt), and density anomaly (σt) qc flag. The GUI can help us to evaluate

instrument drift, verify the integrity of the calibration file, and compute the mean calibration

coefficients and it also offers time-series visualization capability. Visualization is very important

in data quality control: it helps in identifying invalid or doubtful data. Indeed when SST is

plotted, we expect to see of lot of small peaks due to semi-diurnal tidal cycles. We should also

see the typical “bell shape,” indicating the seasonal cycle for an instrument deployed near the

surface (depth < 20 m or so). The visualization step also permits the detection of uncommon

problems like abnormal sensor resolution that cannot be recognized by only computing sensor

drift.

Table 2. Details of temperature sensor models used in the MTN. The column labelled “Period of

Use” gives the range of years over which each model was used. N is the number of times series

recorded with the specified instrument during the 1993–2014 period. There is a grand total of

964 seawater temperature time series recorded during that period. Resolution and accuracy were

provided by the manufacturers in their instrument specifications except for the Vemco Minilog12

whose resolution and accuracy were determined at MLI.

Manufacturer Instrument Model Period of Resolution Accuracy N

Use °C °C

Hugrun Seamon 1993–1999 0.10 0.10 19

Onset Hobo 1994–2002 0.15 0.20 2

RBR TR-1060 2009–2012 0.0005 0.005 6

Sea-Bird SBE-37SI 2002–2014 0.0001 0.002 46

Sea-Bird SBE-37SM 1999–2014 0.0001 0.002 49

Sea-Bird SBE-37SM (+ pressure) 2007–2014 0.0001 0.002 40

Sea-Bird SBE-37SM (+ pump) 2004–2006 0.0001 0.002 5

Sea-Bird SBE-56 2014–2014 0.0001 0.002 9

Stowaway Stowaway 1994–1999 0.13 0.23 4

Vemco Minilog-TR (8 bits) 1995–2007 0.10 0.20 125

Vemco Minilog-TDR (8 bits) 1995–1998 0.10 0.20 5

Vemco Minilog12 (12 bits) 1998–2013 0.01 0.05 461

Vemco Minilog16 (16 bits) 2011–2014 0.01 0.10 84

Vemco Sealog-T 1993–2007 0.12 0.30 109

At the beginning of this program in 1993, we used two types of instruments: Seamon and

Sealog-T. Therefore these instruments are principally associated with the oldest time series of

this program (see Table 2). The Sealog-T had enough space for us to include a very small Hobo

or Stowaway within it as a backup temperature recorder to minimize potential gaps in the time-

series.

The Sealog-T failed six times to record temperature in situations where we were able to use the

backup instrument (two Hobo and four Stowaway; Table 2). Technology evolved, and new

Minilog-TDR and Minilog-TR 8-bit instruments first appeared in 1995, followed by the

Minilog12 in 1998 and the Minilog16 in 2011. Sea-Bird instruments are mainly used on MLI

buoys because of their greater size. A new thermograph support for SBT measurements has

8

recently been built to hold one Sea-Bird SBE-56 instrument. As stated earlier, the support close

to the buoy anchor can receive two side-by-side instruments. Let us now see how this duplication

can be used to verify instrument accuracy. We usually process data from the first instrument. If

this time series shows an acceptable drift then the second time series is just checked and not

processed if the drift is the same or not better than the first one. After this step one instrument is

designated as the main instrument and the second instrument as the backup.

Because of their size and cost, it would have been difficult to install two Sea-Bird SBE-37SM

instruments near the anchor to measure SBT. However, the Minilog12 and Minilog16

instruments are small and cheap, and thus became good candidates to perform duplicate

measurements. Before 2004, the backup file was used only when we had a problem with the

instrument itself or with the raw data. Starting in 2004, instrument drift of the original and

backup instruments were computed and the better of the two instruments (smallest drift) was

kept. The two time series were fully processed by creating raw files allowing us to visualize

them.

We have 76 SBT time-series pairs from 2005 to 2013 that come from 12 stations spread around

the EGSL area. Two of these time-series pairs were eliminated: (1) one time series at Shediac

Valley station in 2009 was withdrawn because the backup instrument showed a resolution

problem, probably due to a weak battery problem, and (2) the time series of the backup

instrument at Sept-Iles in 2006 had a large instrument drift and was discarded. Finally, because

we only wish to compare time series recorded with the same instrument model, we were left with

63 time series pairs of which 51 came from Minilog12 models and 12 from Minilog16 models.

We only focus here on results from the Minilog12. While some may say that using a main and a

backup instrument is doubling the chance of problems, it can also be a very interesting way to

validate data and instrument specifications from the manufacturers.

The distribution of the 417 564 temperature differences (Figure 2) that we have obtained from

the 51 Minilog12 pairs is not Gaussian according to a Kolmogorov-Smirnov test. We found that

different intervals ±0.025°C, ±0.05°C, ±0.10°C, ±0.20°C, and ±0.25°C, represent respectively

83.3%, 93%, 97.4%, 99.1%, and 99.4% of the total 417 564 temperature differences. To

complement this information we checked the interval ±0.5°C which comprises 99.9% of

differences or 417 115 values; thus we still have 449 values outside this interval. The accuracy of

±0.05°C for the Minilog12 was estimated at MLI by Robert Bélanger.

We still have 0.6% of the differences that lie outside the range ±0.25°C with values as large

as -5.4°C (see Figure 3). To gain more confidence in the data recorded by these instruments we

need to take a closer look at these difference values. Minilog instruments are subject to an

important clock drift. This drift is ±2 min/month for the Minilog12 and ±1min/month for the

Minilog16, both of which are very large when compared with a Sea-Bird SBE-56 instrument,

which has a clock drift of merely ±5 s/month. Clock drift may be responsible for some of the

measured temperature differences.

9

Figure 2. Distribution of seawater temperature differences for the Minilog12 for pairs of

instruments deployed at the same depth, location and time period. Only the values in the interval

±0.25°C are shown. There are 415 080 values with a mean difference of 0.00074°C, a standard

deviation of 0.03°C, and a median of 0.001°C.

Most instruments are installed in May and recovered in October, which gives about a 6-month

mooring period. So by September, we could have one Minilog12 with a time lag of 10 minutes

behind while the other Minilog12 could be as much as 10 minutes fast. This means that what we

think are simultaneous measurements for these two instruments could in fact be as much as 20

minutes apart. This large difference in sampling time could be responsible for some of the huge

temperature differences we observed between the main and backup thermographs. Unfortunately,

we do not generally know the true clock drift for these instruments. However, we performed a

test during the download of data in fall 2013: we wrote down the thermograph time and the

computer time to get quantitative estimates of the clock drift of those thermographs. All 17

Minilog12 instruments showed clock advances ranging from 2 to 10 minutes over a period of 5–

6 months. If we accept this instrument behaviour as typical, we can say that the maximum time

difference between two Minilog12 would be 8 minutes after 5–6 months.

10

Figure 3. Seawater temperature differences for the Minilog12 outside the ±0.25°C, interval.

There are 2484 values in this interval, with a minimum of -5.417°C and a maximum of 1.907°C.

The mean difference value is 0.034°C, the standard deviation is 0.463°C, and the median

is -0.264°C.

In an effort to better understand what can cause these differences between the two thermographs,

we will look at the most extreme temperature difference data recorded at station La Perle in

September 2007. We have reported in Table 3a some data from both the main and the backup

thermograph; we can see that at 19:30 UTC we observed a difference of -5.4°C. This is the

largest difference measured in this thermograph comparison among the 417 564 values. It is

worth mentioning that during the previous 24 h the water temperature slowly rose from 4.7°C to

5.6°C, meaning that both instruments were likely in thermal equilibrium with sea bottom

temperature at 18:00 UTC. What is also clear is that both thermographs “saw” the seawater

warming afterwards, but not with the same magnitude. We believe that the main thermograph

measured the water temperature first, just before the warming, and the backup measured the

warming sometime after due to a clock drift. However, if this were the case, the main instrument

should have measured a water temperature close to 12.4°C at 20:00 UTC but instead we have

10.7°C. Moreover, we see a similar temperature difference at the next sampling time. To explain

11

this, we think that the main instrument was probably buried in the mud (one of us [R.D.]

occasionally saw instruments in the support bracket coming back from the field with one

instrument having a mud cap) while the backup had better contact with seawater. Perhaps the

presence of mud increased the response time constant of the main thermograph, and so it took

longer for it to measure the water warming or cooling compared with the backup instrument.

Regarding the temperature sensor time response constant, we can add that a sensor does not

immediately react to a sudden temperature change. It must first adjust to this temperature change

before measuring correctly. For a Minilog12, the manufacturer specifies a time response constant

of 30 seconds in stirred water.

Table 3a. Some instantaneous seawater temperatures recorded by the main and backup

instruments at La Perle station on 5 September 2007 at 26 m.

Time (UTC) Main (°C) Backup (°C) Difference (°C)

2007-09-05 18:00 5.586 5.554 0.032

2007-09-05 18:30 5.630 5.610 0.020

2007-09-05 19:00 5.652 5.643 0.009

2007-09-05 19:30 6.954 12.371 -5.417

2007-09-05 20:00 10.724 13.038 -2.314

2007-09-05 20:30 11.590 13.007 -1.417

2007-09-05 21:00 11.423 12.517 -1.094

2007-09-05 21:30 10.807 9.914 0.893

2007-09-05 22:00 10.031 11.347 -1.316

2007-09-05 22:30 10.776 11.734 -0.958

2007-09-05 23:00 11.340 12.215 -0.875

Table 3b. Some instantaneous seawater temperatures recorded by the main and backup

instruments at La Perle station on 9 August 2007 at 26 m.

Time (UTC) Main (°C) Backup (°C) Difference (°C)

2007-08-09 16:30 8.962 9.031 -0.069

2007-08-09 17:00 8.962 9.031 -0.069

2007-08-09 17:30 8.973 9.042 -0.069

2007-08-09 18:00 9.397 10.157 -0.760

2007-08-09 18:30 14.720 15.784 -1.064

2007-08-09 19:00 14.853 15.390 -0.537

2007-08-09 19:30 14.658 15.080 -0.422

2007-08-09 20:00 14.412 14.666 -0.254

2007-08-09 20:30 13.980 14.179 -0.199

2007-08-09 21:00 13.784 13.951 -0.167

We have reported in Table 3b a similar behaviour that occurred a month earlier and probably

confirms that the instruments were not malfunctioning. Of course we cannot cover in detail here

our investigation of the 2484 values found outside the ±0.25°C interval. Issues that we

considered included trying to identify if there was an electronic problem (like battery failure), or

rather a rapid change in SBT that is not seen at the same time by the two instruments because of

instrument clock drift or a time response constant that was modified by the presence of mud or

other obstructing material, resulting in seawater temperature differences between the two

12

thermographs records. The Minilog12 is a good and inexpensive instrument. However we must

take into account possible instrument clock drift when comparing time series.

Taking into consideration the 417 564 temperature differences that we obtained from the 51 pairs

of Minilog12 instruments, we obtained a mean temperature difference of -0.000381°C, a

standard deviation of 0.0469°C, and a median value of 0.001°C.

In Figure 4 we plotted the 111 072 temperature differences obtained from the 12 pairs of

Minilog16; the values range roughly within the [-1.1°C, 1.2°C] interval. The temperature

differences from counts between about 76 250 and 103 000 come from thermographs that were

measuring SBT at 330 m depth at Rimouski station during field seasons 2011, 2012, and 2013.

Because temporal seawater temperature variations are very weak at this depth, thermograph

clock drift cannot induce large temperature differences.

Figure 4. Seawater temperature differences for the Minilog16. There are 111 072 values, that

come from 7 stations and 12 pairs of instrument, recorded from 2011 to 2013. The mean value

difference is 0.002°C, the standard deviation is 0.043°C, and the median is 0.001°C.

13

2.4 INSTRUMENT DEPTH ADJUSTMENT AND OTHER RELEVANT DETAILS

The nominal instrument depth for SST (Figure A1.3) is probably more reliable than the nominal

instrument depth used for SBT. The actual instrument depth for SST is the distance between the

thermograph support on the buoy and the (mean) waterline, which remains constant. The actual

instrument depth for SBT should be about equal to the water depth, with some uncertainty that is

worth mentioning; indeed, only the instrument depth for IML, IML PEM, and Borden stations

are already adjusted in Table 1. Once the buoy tender is on station for a CCG buoy deployment,

the instruments are put inside the support and attached with a shackle directly to the ring on top

of the buoy anchor (Figure A1.4). However, due to buoy movement, the anchor will eventually

tip over and lay down on its side so that finally the instrument depth corresponds to water depth.

We suppose that most of the time the anchor is not upright. But in circumstances where the

anchor is upright, instrument depth would be about L metres less than the local water depth,

where L is the CCG buoy anchor height.

A CCG buoy anchor is specially made to minimize buoy displacement, with spikes on its side to

hinder movement. However, if the anchor is forced to move, it will plough the seafloor, building

up a pile of mud (or sand) and probably bury the instruments support. Since 2014, in order to

minimize the possibility of buried instruments, which can increase the thermograph time

response constant as discussed above, we have installed the shackle on the chain at 1 m from the

anchor, hoping that the instrument will stay in the water even when the anchor lays down on the

seafloor. In this case, if we consider the anchor to be upright, then the instrument depth would be

about L + 1 m less than the water depth. Because the CCG buoy anchor height and weight

depend on the size of the CCG buoy, Table 4 reports the anchor height and weight for the CCG

buoys used in the thermograph network.

Table 4. Height (L) and weight of CCG buoy anchors used at thermograph network stations.

Station name is followed by sea bottom type in parenthesis, where MB, RB, and SB mean

respectively muddy bottom, rocky bottom and sandy bottom. The Grande-Rivière station was

equipped with a CCG buoy until 2011.

Station L (m) weight (kg) Station L (m) weight (kg)

Baie-Comeau (MB) 1.5 3636 La Tabatière 0.75 1818

Bic (RB) 1.1 2955 Natashquan (SB) 0.5 682

Blanc-Sablon (RB) 1.1 2727 Port-Menier (RB) 1.1 2727

Grande-Rivière 1.5 3636 Riv.-au-Tonnerre (MB) 1.1 2727

La Perle (SB) 1.1 2727 Sept-Iles (SB) 1.1 2727

La Romaine (RB) 0.9 2273 Tadoussac (RB) 1.1 2727

In 2009, we learned that most CCG buoys are not installed right on the shoal but are instead put a

short distance away from it. This caused a difference in the water depth instrument that we had

initially inferred and the real one. We used to deduce the water depth directly from chart datum.

We made changes in the raw file header on 8 an 9 July 2009, using information available through

the CCG’s SIPA database (this is a French acronym that is also used in English and means

“Système d’Information sur le Programme des Aides”). We also changed the water depth

metadata at Shediac Valley station on 7 January 2016. The water depth at IML station was first

14

changed from 13.5 m to 15.0 m in July 2009, then on 8 January 2016 from 15.0 m to 14.0 m. All

the changes are noted in Table 5.

Table 5. Water depth corrections. The station name is followed by the former depth and the

corrected depth values. Corrections in the raw file database were carried out on 8–9 July 2009

and 7–8 January 2016.

Station Former

Depth (m)

Corrected

Depth (m)

Station Former

Depth (m)

Corrected

Depth (m)

Baie-Comeau 80.0 82.3 La Tabatière 36.0 39.0

Banc Beaugé 100.0 97.0 La Romaine 14.0 21.9

Bic 5.8 30.5 Natashquan 7.5 5.4

Grande-Rivière 7.0 10.0 Sept-Iles 25.0 21.9

Havre-St-Pierre 100.0 120.0 Shediac Valley 82.0 85.8

IML 15.0 14.0 Tadoussac 30.0 36.6

La Perle 8.5 26

When available, a direct measure of instrument depth can be derived from thermograph pressure

sensor data. Otherwise, it must be deduced from water depth (H) and instrument height (h) from

the seafloor as

Z = H – h (1)

The best available estimate of water depth comes from the SIPA database, followed by estimates

from chart datum. The bottom instrument depths, Z, in Table 1 are the nominal values that are

often set to water depth as a first approximation. Instrument depth for a thermograph installed

close to a CCG buoy anchor is obtained using equation 1, with h set to the CCG buoy anchor

height, L, before 2014 and L + 1 since 2014. Next instrument heights (h) are described for

various stations:

Banc Beaugé. From 1998 to 2011, the Banc Beaugé station was equipped with a first

generation MOB (Figure A1.6) called IML-2. Starting in 2012, an MLI buoy smaller than

a first generation MOB was used. The anchor consists of two train wheels, so the

thermograph is about 0.45 m above the seafloor. The bottom material is not known.

Belle Isle. The mooring at the Belle Isle station replaced the one at the Old Belle Isle

station in November 2007. The water depth was deduced from chart datum. Estimated

water depth can vary from 70.2 to 77.0 m while distances from the bottom to the

instrument vary from 4.5 m to 6.8 m. However, the SBT is measured with a SBE37-SM

equipped with a pressure sensor, meaning that instrument depth is accurately known for

every SBT value.

Bonne Bay. The Bonne Bay station mooring was installed and recovered by a person

under contract with MLI. In 1995 we used a Minilog-TDR to record SBT and pressure;

the 1995 depth time series is shown in Figure 5. The accuracy of the depth value is about

±2 m. It seems that the mooring drifted into shallow water until the field worker

repositioned the mooring to the correct location. Because the thermograph was very close

15

to the seafloor, the instrument depth was likely within the [18.0 m, 22.0 m] interval while

in the correct position.

Borden. The Borden station is the shallowest station of the network measuring SBT.

During the period from 19 July 2010 to 27 November 2010, an SBE-37SM with a

pressure sensor was used. We measured a minimum depth of 1.6 m and a maximum

depth of 4.2 m. Just before instrument mooring in July 2010, the pressure sensor showed

a pressure offset of 0.35 m, giving a minimum depth of 1.25 m. Because instrument

height is about 0.25 m above the seafloor we estimate that the water depth (H) is about

1.5 m.

Note: Instrument depth from 1995 to 2002 was 1.5 m.

Courant de Gaspé. The Courant de Gaspé station is equipped with a first generation

MOB called IML-7. The buoy is exposed to strong currents that forced MLI staff to

adjust the weight and shape of the buoy anchor to minimize displacement. Since the

beginning in 2005 until 2014, we used two train wheels as a buoy anchor. We thought

that a first buoy drift event in 2007 was exceptional and so we did not change the buoy

anchor design (more precisely we had a first drift during 30 June–1 July, about 0.5 km to

the North and 1.8 km to the East then we had a more important drift on 28–29 October

where the buoy drifted about 3.2 km to the North and 16.7 km to the East). We had

another buoy drift event in 2013 (the buoy drifted about 0.4 km to the North and 1.4 km

to the East on 11 August), so we modified the buoy anchor in 2014 to three train wheels

sitting on a metal plate (Figure A1.5). When large tensions are applied on the anchor

cable the anchor will tilt and the metal plate will dig into the mud. In cases where the

anchor was two train wheels, the instrument is 0.45 m above the seafloor, whereas the

set-up of three train wheels on a metal plate leaves the instrument 0.75 m above the

seafloor. Because mud is present, it seems reasonable to subtract 0.3 m; thus the

instrument was likely about 0.15 m off-bottom with the two-train-wheel anchor and about

0.45 m above seafloor for the three-train-wheel anchor.

Grande-Rivière. The Grande-Rivière station was equipped with a CCG buoy from 1994

to 2011. Since 2012, the station has been equipped with a smaller buoy called IML-9.

The anchor consists of one train wheel, so the instrument is about 0.30 m above seafloor.

The seafloor material is not known.

Gyre Anticosti. The Gyre Anticosti station was equipped with a first generation MOB

called IML-8. The anchor was two train wheels, meaning that the thermograph was about

0.45 m above seafloor. The bottom material is not known.

Havre-St-Pierre. The Havre-St-Pierre station is equipped with an MLI buoy called IML-

1. The anchor is one train wheel, so the thermograph is about 0.30 m above seafloor. The

bottom material is not known.

Hudson-1. The Hudson-1 station is a mooring. We started to measure SBT in fall 2013.

The thermograph is about 2.3 m above seafloor including the one-train-wheel anchor.

The bottom material is not known.

Hudson-3. The Hudson-3 station is a mooring. During the mooring in September 2012, a

depth sounding gave a water column depth of 161.7 m while during the mooring in

October 2013 a depth sounding measured a water depth of 162.0 m, which is nearly

identical. The thermograph is about 2.3 m above seafloor including the one-train-wheel

anchor. The bottom material is not known.

16

IML. The IML station is equipped with a small MLI buoy called IML-3. The anchor is

one train wheel, so the thermograph is about 0.3 m above seafloor and instrument depth

is about 13.7 m. The bottom material is not known.

IML PEM. The IML PEM (PEM is a French acronym meaning: Prise d’Eau de Mer)

station has its thermograph attached to the water intake of MLI’s seawater pumping

system. For the last three years, the SBE-37SM that is equipped with a pressure sensor

has allowed us to measure the instrument’s depth at about 12.4 m. We have estimated

that H is about 13 m.

Ile Shag. The Ile Shag station is equipped with a very small surface buoy for SST

measurements. SBT is measured with a thermograph on a mooring line lying on the

seafloor. From 21 September 2007 to 23 May 2008, we used an SBE-37SM that was

equipped with a pressure sensor. We measured a minimum depth of 9.44 m and a

maximum depth of 11.48 m. Just before instrument mooring, the pressure sensor showed

a pressure offset of -0.11 m. We should thus adjust the minimum depth to 9.55 m;

however, the thermograph is about 0.2 m above seafloor, giving an instrument depth of

about 9.75 m.

Irving Whale. The Irving Whale station is equipped with an MLI buoy called IML-5.

The anchor is one train wheel, meaning that the thermograph is about 0.30 m above

seafloor. The bottom material is not known.

Mont-Louis. The Mont-Louis station is equipped with an EC meteorological buoy

(Figure A1.6). During the 1994–1995 field season, a mooring was installed about 660 m

from the EC buoy at a water depth of about 334 m. The thermograph was inside the

acoustic release gear that is about 1 m above the seafloor. The thermograph at the

intermediate (200 m) depth is installed in late fall and recovered the next spring.

According to pressure data recorded in 2010, the instrument depths varied from 192.7 m

to 197.4 m, with a mean of 194.5 m. It is worth mentioning that the EC buoy position has

changed a little over the years. This is mainly due to fishing gears that get entangled with

the EC buoy anchor line. The buoy positions through the years are presented in Table 6.

Old Belle Isle. At the Old Belle Isle station, a mooring recorded year-round from January

2004 to November 2007. The water depth was deduced from chart datum since the

thermograph had no pressure sensor. The instrument was about 7.3 m above seafloor,

which gave 103.7 m for the instrument depth. The anchor was not recovered and the

bottom material is not known.

Rimouski. The Rimouski station was equipped with a first generation MOB from 2002 to

2012 called IML-4. During that period the buoy anchor was two train wheels, so the

thermograph was 0.45 m above the seafloor. Because the sea bottom was known to be

mud, it seems reasonable to subtract 0.20 m, so that the instrument is likely 0.25 m above

the seafloor. Since 2013, the station has been equipped with a Viking MOB with three-

train-wheel anchor (Figure A1.5), meaning that the thermograph is about 0.75 m above

the seafloor. Because of the mud, in this case we subtract 0.3 m, so the instrument is

likely 0.45 m above the seafloor. The IML-4 buoy once drifted about 350 m to the East

on 22 September 2002.

Shediac Valley. The Shediac Valley station was equipped with a first generation MOB

from 2004 to 2010 called IML-6. During that period, the buoy anchor was two train

wheels, meaning that the thermograph is about 0.45 m above the seafloor. Because the

sea bottom was known to consist of mud, it seems reasonable to subtract 0.2 m. This will

17

give a thermograph that is about 0.25 m above the seafloor. In 2011, the recovery buoy’s

handle broke during loading onto the ship, so we cancelled the 2011 monitoring season.

During the field seasons 2012 and 2013, the station was equipped with an MLI buoy with

a one-train- wheel buoy anchor, meaning that instrument depth is very close to water

depth. In 2014 we re-equipped the station with a first generation MOB.

Tadoussac. A mooring was deployed from November 2002 to June 2003 at 48° 7.3 N,

69° 40.3 W. The water depth is about 45 m according to the chart datum and the

instrument depth is about 43 m. This station was named Tadoussac-exp (43) and was

about 340 m north-east of the Tadoussac station listed in Table 1.

Figure 5. Instrument depth at Bonne Bay station.

Table 6. Positions of the EC meteorological buoy at Mont-Louis station from 1994 to 2014.

Starting year Ending year Latitude N Longitude W

1994 1994 49° 33.618 65° 45.552

1995 1995 49° 31.998 65° 43.698

1996 1997 49° 32.808 65° 44.508

1998 2003 49° 32.592 65° 46.320

2004 2004 49° 33.210 65° 45.798

2005 2005 49° 33.270 65° 45.528

2006 2013 49° 32.598 65° 45.582

2014 ongoing 49° 32.346 65° 42.642

18

2.5 CHANGE IN STATION POSITION

We wish to emphasize the importance of keeping stations at fixed positions if at all possible.

Suppose someone would like to use SST time series from the MTN for a study on climate

change in the EGSL. Before proceeding, this person would need to make sure that the station has

never been relocated.

Indeed, the time-series statistics can be affected if a station is moved, so observed changes

cannot be ascribed only to climate variability but also to some effect related to relocation of

station (Vincent, 1998).

Figure 6. Seawater temperature differences between the original Grande-Rivière site and the

proposed Grande-Rivière site at 2 m depth.

If the station displacement is permanent, it should be well documented, and we should (at least

ideally) have one season of overlapping data to compare time series and adjust the new one. In

19

2011, we were forewarned by CCG that a buoy at Grande-Rivière station was going to be

relocated by an alignment on the Grande-Rivière pier for season 2012.

To prepare ourselves for this change, during season 2011, in addition to the buoy installed at the

original Grande-Rivière station, another buoy was moored on the proposed new site of Grande-

Rivière station that is located on the same isobath as the original station, but about 250 m to the

north. In Figure 6, we show the large SST instantaneous differences that vary from -2.2°C to

5.2°C and the mean SST daily differences varying from -0.14°C to 1.2°C between the two sites

at 2 m depth. We also found monthly mean differences of 0.09°C, 0.17°C, 0.12°C, and 0.02°C

for the months of June to September 2011 respectively. Monthly mean differences about equal to

or greater than 0.1°C are important, and adjustment on SST time series from the old site to the

new site must be done when one attempts to use it in climatological studies. These statistics

clearly show that even a small displacement of a station can have important effects on time-

series statistics that are not totally related to climate variability and change. In the end, the CCG

agreed to keep the Grande-Rivière station at the original location that we had been using since

1994.

3.0 RESULTS

Statistical results from the thermograph network such as monthly means of water temperature

and salinity and plots of daily water temperature and salinity are routinely published in annual

DFO research documents on the physical oceanographic conditions of the Gulf of St. Lawrence

(e.g. Galbraith et al., 2015). As a complement to the results already available, missing records

will be documented here, and examples of oceanographic processes present in the EGSL will be

shown.

3.1 MISSING DATA ESTIMATION

As of the end of 2014, there were 964 time series and this number is expected to exceed 1000 in

2016. When one wants to use these data, it can be very helpful to know if data from a specific

station for a specific period of time have been recorded or not or partially recorded.

Thermographs are electronic devices that can fail to measure water temperature, creating gaps

that can be short or long. In some other cases, instead of missing records, we will have

unscheduled supplementary records that only occur due to the MOB controller and are likely

caused by interference between the buoy data acquisition system and online communication done

by MLI staff for operational purposes. During the visualization step of time series, doubtful or

erroneous data are sometimes detected and flagged, increasing the number of missing data.

As a last effort to qualify time series we have computed some record numbers, described below,

that summarize the extent to which a temperature series is complete. For this purpose, we used

information contained in the header of TS9 files, such as the times of first and last valid record

and the sampling interval, to compute the expected number of records, Ne. This is the number of

samples that we should normally get, and it serves as a reference to evaluate the number of

missing records.

20

A computer program reads valid data from the time of the first valid sample to the time of last

valid sample; this gives Nv(t), which is the date and hour of every valid record. The number of

valid records Nv is the size of Nv(t). It is worth mentioning that the word “valid” includes all

records, even those with a quality control flag (Table 7) greater than 1. Let Nf(t) be the time

series of the date and hour of flagged records having values of 2, 3, 4, or 9, with a length of Nf.

Table 7. Quality control flag meanings.

Flag Meaning

0 No quality control was done

1 Datum seems correct

2 Datum seems inconsistent with other data

3 Datum seems doubtful

4 Datum seems incorrect

5 Datum has been modified

6 to 8 Undefined

9 Datum is missing

To obtain the number of missing records, Nm, normally we just have to subtract Nv from Ne.

However, Nv is sometimes greater than Ne, leading to a negative value for Nm. This implies that

we have read more values than expected. To take this oversampling issue into account, we can

write down the following relation:

Ne – Nv = Nm – No (2)

where No in the number of oversampled records.

In equation 2, even if Ne – Nv is zero, Nm and No may still be non–zero. To determine Nm and

No, the number of good records is introduced. A good record is a single record taken within ± 3

minutes of the estimated time. Any record outside this time interval is considered to be

oversampling. We will also consider as an oversampling record the case of successive records

that are separated by less than 3 minutes. The number of valid records can then be defined as:

Nv = Nf + No + Ng (3)

We can combine equations 2 and 3 to get an equation for Nm that is independent of No:

Nm = Ne – Nf – Ng (4)

We will define a final number that is the total number of missing records:

Ntm = Nm + Nf (5)

Even if the proposed missing data estimation model defined by equations 2–5 is general, the way

we compute the various record numbers is not foolproof, meaning that an oversampling record

21

would sometimes be better described as a delayed record; an example of this issue is shown in

the last paragraph of section 4.1. The number of missing records does not tell the full story, and

the localization of data gaps can be very important. For example, large gaps can render time

series unsuitable for harmonic analysis. For every temperature time series from 1993 to 2014, we

have computed Ne, Ng(t), No(t), and Nm(t), and the resulting numbers, including Nf and Nv, are

presented in tables in Appendix 2. We have also included two other numbers derived from Ng(t)

in these tables: the number of gaps, NOG, and the maximum gap, MG. In these tables the suffix

“exp” which stands for experimental, has been added to some station names. Experimental

measurements can occur for various reasons (e.g., specific demands from researchers, depth

validation, diurnal warming study, moving a station) and are situations where an instrument is

added to an existing station for a restricted period of time that can last some years; this

instrument will eventually be removed once the experiment ends. Table 8 contains definitions of

various record numbers.

Table 8. Record number definitions.

Number Definition

Ne Expected record number

Nv Valid record number

Nf Flagged record number

Nm Missing record number

No Oversampling record number

Ng Good record number

Ntm Total number of missing record

NOG Number of gaps

MG Maximum gap

Here is an example to show how the Appendix 2 tables are constructed. If we look at Table

A2.22 at Old Harry station, which is equipped with a Viking MOB called IML-10, we should

have had 13 481 records according to the Ne value. On the other hand according to the Nv value,

we have read only 10 194 records, 10 191 of which are good data (Ng) and 3 of which are

flagged data (Nf). So all the 10 191 good data have been recorded at the right time. In

accordance to relations (4) and (5), the total number of missing records Ntm = Nm + Nf = Ne –

Ng = 13 481 – 10 191 = 3290. Because we have three flagged data with Nf = 3, we get Nm =

3287. We also see that we have 82 gaps and the maximum gap has a size of 3208 records. Apart

from the MG value, that still leaves 82 missing records distributed among the other 81 gaps. This

huge gap of 3208 records was caused by a battery failure that required MLI staff to visit IML-10

to fix the problem.

In the Appendix 2 tables, we can observe irregularities if one attempts to compute the number of

missing records using relation (2). To better understand this issue, we will proceed with an

example by choosing the Bic (1) station from Table A2.3. We should have Ntm = 1 instead of

Ntm = 0, because Ne – Nv = 7658 – 7657 = 1. By looking over the raw file associated with these

data, we realized that approximately once every 24 hours the sampling time interval between two

consecutives records is 8 seconds longer. Because this 8 seconds delay is much shorter than our

±3 minutes criterion for sampling time, we could not detect any missing records. Globally, this is

equivalent to having a longer sampling interval, resulting in fewer records than expected.

22

According to the manufacturer, this problem is likely due to a weak battery problem. We

observed this same technical issue at several other stations; these are listed in the following

tables: Table A2.3: Bonne Bay(25), La Romaine (1), and Mont-Louis(30); Table A2.4: Mont-

Louis (30); Table A2.6: Irving Whale(67), La Perle (26), and Tadoussac(36.6); Table A2.7:

Bonne Bay (0.5); Table A2.9: IML PEM(12.4); Table A2.10: Borden(1); Table A2.15: IML

PEM (12.4). These entries were written in bold type to facilitate their detection.

3.2 YEARLY DATA AVAILABILITY CHART

Though the data metrics presented in the previous section provide a good overview of the water

temperature records as a function of time, we also present a more visual way of showing all the

data recorded during a specific year from 1 January to 31 December for all stations. We used the

preceding results that served to compute the record numbers described in tables A2.1–A2.22 to

build a yearly data chart for every thermograph network station. Because some stations record

year-round, we have to merge data from different instrument deployments to display all available

data from a specific year. This can cause some apparent discrepancies for year-round stations

between the results presented in Appendix 2 tables and the numbers that have been written for

some stations in Appendix 3 figures.

In these figures, the station name is followed by the record numbers No, Ntm, Nf, and MG.

Because these four numbers are generally equal to zero, they are not indicated except when Ntm

or No is non-zero. This facilitates the identification of time series having issues in terms of

record numbers. For each station, we plotted a continuous horizontal line showing the range of

valid data records for that year, then we drew the locations of oversampling records, the total

missing records, the flagged data with quality control flag greater than 1, except the qc flag equal

to 5, and the maximum gap. In cases when we have more than two equal MG values, only the

first one is plotted.

3.3 SPRING–NEAP TIDAL CYCLE AT BIC STATION

Among the seawater temperature time series recorded by the MTN since 1993, it seems those

measured at Bic station are the most sensitive to the spring and neap tidal cycle. The 2007 Bic

station water temperature time series is perhaps the best candidate to illustrate how seawater

temperature variability can be largely modulated by the astronomic events of full moon and new

moon (see red curve in Figure 7, where semi-diurnal and diurnal tides are filtered out to enhance

the spring and neap tidal signal).

Two interesting delays are readily apparent in Figure 7. The first delay, referred to as “age of the

tide” in oceanography, is the delay of about 1 to 2 days between the astronomical full moon and

maximum daily tidal range. The second delay is that between maximum tidal range and

minimum sea surface temperature. Tidal currents are at their maximum speed when tidal height

range is maximum. Those maximum tidal currents favor various phenomena (internal tides,

solitary waves, mixing through shear instabilities) that can entrain cold intermediate layer (CIL)

23

waters into the surface layer. The spring-neap SST minima generally occur 2.5 to 8 days after the

full moon or the new moon.

Figure 7. Temperature at Bic station at 1 m depth in 2007. The blue curve shows all data while

the red curve is using a 48-hours triangular filter. The black full circles indicate the new moon

while the empty circles indicate the full moon. The vertical dotted lines pinpoint the local non-

filtered SST minimum for each spring-neap tidal cycle. The lower black curve is the hourly

water level (divided by 2 to improve graphic visualization) measured at Rimouski by a Canadian

Hydrographic Service tide gauge (station no. 2985).

3.4 WIND EFFECT AT LA TABATIERE STATION

It is well known that coastal waters of the Québec Lower North Shore are subject to episodic

upwelling (Bourque and Kelley, 1995) and downwelling events induced by wind. The La

Tabatière station data show two major downwelling events in 2014 (Figure 8, [top]) that coincide

with northeast wind events measured at Blanc-Sablon (Figure 8, [bottom]), 100 km away, by the

Atmospheric Environment Service (AES, Environment Canada).

24

Figure 8. Seawater temperature at La Tabatière station in 2014 at 1 m (blue dotted line) and 39 m

(black line). The black full circles indicate the new moon while the empty circles indicate the full

moon (top). Daily wind measured at Blanc-Sablon in 2014 by AES. The wind direction is

defined here as the direction toward which the wind blows. Only winds higher than 32 km/h are

shown (bottom).

There are gaps in the time series of daily winds caused by winds lower than 31 km/h. In Figure

8, a downwelling event was observed to begin around 8 August and ended on 17 August. It

25

appears likely that northeast wind at the beginning of August could explain this first

downwelling event. The comparison between the wind and the second downwelling event, which

started on 20 August to end on 25 August, is less conclusive.

4.0 DISCUSSION

4.1 THE OVERSAMPLING ISSUE

The oversampling issue discussed in section 3.1 was discovered with the deployment of MOB in

the EGSL, and is exclusively found in those buoys. Every 15 minutes the MOB controller issues

about forty commands that last less than 3.75 minutes. The pump is started to clean the tubing,

then optical measurements are made for one minute followed by some statistics computing, then

meteorological sensors and oceanographic sensors measure also for one minute. Finally the

controller sends data according to its configuration. Roughly speaking, optical data streams end

at minutes 0, 15, 30 and 45 while seawater temperature and conductivity data streams end at

minutes 3, 18, 33, 48; this is what we usually get but not exclusively.

Only one MOB (Rimouski station) transmits every 15 minutes using a UHF link while other

MOBs transmit a chosen group of parameters every 4 hours using a satellite link, although that

used to be every 8 hours, with only a subset of data recorded at minute 3 transmitted to save on

telecom costs. We do not think that the problem of oversampling is linked to instrument

measurements themselves as it is actually the MOB controller that sends commands to the

thermograph (Sea-Bird-37SI) to take measurements at every sampling interval (usually 15

minutes). This problem does not appear to be directly related to satellite telecom either. It might

be interference between the data acquisition system and online communication with the buoy by

MLI staff. It seems that the controller waits for a signal before proceeding with data acquisition

and consequently misinterprets the MLI staff communication as a signal to start data acquisition.

In 2014, Courant de Gaspé (0.5) and Rimouski (0.5) stations both had one record from

oversampling (No=1). At Courant de Gaspé station on 28 July, data were recorded at 00:03:22,

00:18:22, 00:33:22, 00:48:22 and 00:51:44 indicating the presence of oversampling at 00:51:44.

At Rimouski station on 15 May data were recorded at 22:03:24, 22:18:24, 22:33:23, 22:42:25

and the next record is at 2014/05/18 18:03:27. The oversampling record at 22:42:25 is just

preceding the max gap (MG = 269 records). The oversampled water temperature value appears

to be correct when compared with surrounding values. The detection of oversampling records is

not irrefutable, but indicates possible problems with the recording.

We provide a last example about oversampling records by looking at results from Shediac

Valley (0.5) station in 2008 (Table A2.16). The usual way to determine the number of missing

records in a time series is to subtract Nv from Ne; in this example Ne – Nv = 16 693 – 16 693 =

0, implying that no record is missing. According to equation (2) even if Ne – Nv is zero, it does

not necessarily follow that Nm (or No) is zero. Here we have Ne – Nv = 0 because the number of

missing records is exactly compensated by the number of oversampling records.

26

We mentioned earlier that the oversampling issue is only related to MOB. However, we have

found one exception to this rule at station IML (0.5) in 1999 (Table A2.7). The SBE-37SM was

programmed to take a sample every 5 minutes, but recorded on 26 November 1999 16:27:30

instead of 26 November 1999 16:25:00. Technically speaking we have here a real oversampling

record, however we can also say that it was not an extraneous oversampling record, but simply

recorded 2.5 minutes later than expected. This has raised the number of missing record by 1 but

the other 6 missing records are real.

4.2 THE MISSING RECORD MODEL

To characterise the missing records in water temperature time series we have presented some

record numbers that are linked together by four mathematical relations (section 3.1). We have

gathered in Tables of Appendix 2 all the computed numbers for every raw file in our database.

Some stations maintained year-round can have more than one table entry. Recall that record

numbers Nv, No, Ng, Nm and Nf are the size of time series Nv(t), No(t), Ng(t), Nm(t) and Nf(t)

respectively. The time series Nv(t) and Nf(t) are readily obtained during the file reading step

whereas the other three time series need specialized algorithms and computer programs to be

estimated. Using these record number time series, yearly data charts were produced for every

year, resulting in 22 plots in Appendix 3. Together these figures form a kind of yearly catalog of

available water temperature time series.

5.0 CONCLUSIONS

Since 1993 we have strived to build a thermograph network in the Gulf of St. Lawrence

dedicated to collect sea surface temperature (SST) records. As this program evolved sea bottom

temperature (SBT) also began to be recorded, instruments with conductivity cells were used and

oceanographic buoys with real-time data transmission capability were developed in order to

improve our knowledge of the oceanography of the EGSL.

In section 2 we described the nuts and bolts of our data quality control. Our results suggest that

we should privilege the use of low clock drift instruments, such as the Sea-Bird SBE-56 that has

a ±5 s/month clock drift, because problems associated with clock drift are difficult to detect. We

have also shown that moving a “fixed” station must be done with great care if it cannot be

avoided. At least one entire field season of overlapping data is necessary to describe the

transition from the old site to the new site.

The relatively large annual volume of data can justify creating software tools that enable a rapid

look at the data to detect issues about seawater temperature time-series record numbers. We

presented a homemade record number model to estimate and locate missing records and other

pertinent record numbers. We produced tables containing record numbers for every raw file, and

then using time series record numbers we plotted yearly data charts. These plots provide a quick

and convenient way of identifying available data for all stations, which together with the record

number tables constitute the main results of this report.

27

We have always considered that the MOBs form a separate network within the MTN. This is due

to the two-way, real-time data transfer and the multi-device and multi-sensor platform capability

that make them unique. Real-time in situ SST data are extremely useful: they can be used to

validate satellite infrared sensor data from the well-known AVHRR or data from the more recent

satellite microwave sensors. Similarly, real-time sea surface salinity (SSS) data can help calibrate

satellite SSS measurements, such as those from the Soil Moisture and Ocean Salinity (SMOS)

mission of the European Space Agency (ESA, ongoing) and the Aquarius/SAC-D (mission starts

10 June 2011, ends 17 June 2015) of the National Aeronautics and Space Administration

(NASA) in collaboration with Argentina.

Finally, we hope this report will be useful for the future of the Maurice Lamontagne Institute’s

thermograph network.

6.0 ACKNOWLEDGEMENTS

During the 22 years on which we are reporting here, many people have contributed to

maintaining the thermograph network, especially those who have taken care of instrument

calibrations; these include Robert Bélanger (now working for Defense Research and

Development Canada, Valcartier, QC), Daniel Thibeault (now working at the CCG College,

Sydney, NS) and Michel Rousseau of Science Advice, Information and Support Branch

(SAISB). We acknowledge the great contribution of Roger Pigeon (SAISB) to the conception

and the building of the MLI oceanographic buoys (MOB). We are grateful to Daniel Beaulieu

(now retired) for building diverse robust thermograph supports and to CCG staff for their

contribution to the MTN. The authors thank Laure Devine (technical report editor, SAISB) and

the reviewers Peter Galbraith of Pelagic and Ecosystem Science Branch (PESB) and Diane

Lavoie (PESB) for their constructive comments. We also thank people under MLI contract for

their past and present contribution to this program: Raymond Arsenault, Roy Bannister, Jean-

Yves Couture, Félix Saint-Pierre, and Martin Turbide.

7.0 REFERENCES

Bourque, M.-C., and Kelley, D.E. 1995. Evidence of wind-driven upwelling in Jacques-Cartier

Strait. Atmosphere-Ocean 33 (4): 621–637.

Galbraith, P.S., Chassé, J., Nicot, P., Caverhill, C., Gilbert, D., Pettigrew, B., Lefaivre, D.,

Brickman, D., Devine, L., and Lafleur, C. 2015. Physical oceanographic conditions in the

Gulf of St. Lawrence in 2014. DFO Can. Sci. Advis. Sec. Res. Doc. 2015/032. v + 82 p.

GCOS-164. 2012. Report of the Twentieth Session of the WMO-IOC-UNEP-ICSU Steering

Committee for GCOS, Geneva, Switzerland, September 2012, Doc. 16.2, GCOS 20th

Anniversary ‘Think Tank’ Session.

28

Gilbert, D., Vézina, A.F., Pettigrew, B., Swain, D.P., Galbraith, P.S., Devine, L., and Roy, N.

1997. État du golfe du Saint-Laurent : conditions océanographiques en 1995. Rapp. tech. can.

hydrogr. sci. océan. 191: xii + 113 p.

Pettigrew, B., Larouche, P., and Gilbert D. 2011. Validation des images composites des

températures de surface produites au laboratoire de télédétection de l’Institut Maurice-

Lamontagne. Rapp. tech. can. hydrogr. sci. océan. 270: viii + 31 p.

Saucier, F. J, Roy, F., Gilbert, D., Pellerin, P., and Ritchie, H. 2003. Modeling the formation and

circulation processes of water masses and sea ice in the Gulf of St. Lawrence, Canada. J.

Geophys. Res., 108, 3269, doi: 10.1029/2000JC000686, C8.

Therriault, J.-C., Petrie, B., Pépin, P., Gagnon, J., Gregory, D., Helbig, J., Herman, A., Lefaivre,

D., Mitchell, M., Pelchat, B., Runge, J., and Sameoto, D. 1998. Proposal for a Northwest

Atlantic zonal monitoring program. Can. Tech. Rep. Hydrogr. Ocean Sci., 194: vii + 57 pp.

Vincent, L. A. 1998. A technique for the identification of inhomogeneities in Canadian

temperature series. J. Climate 11: 1094–1104.

29

APPENDIX 1. THERMOGRAPH NETWORK PHOTO LIBRARY

Figure A1.1 Instruments used in the thermograph network. From left to right: 45.7 cm long

ruler for scale, SBE-37SM, SBE-56, TR-1060, Minilog16, Minilog12, Minilog8, Stowaway,

Sealog-T, and Hugrun. At the extreme right is an electro-mechanical Ryan thermograph model

that was used before the first digital thermographs such as the Hugrun and Sealog-T.

30

Figure A1.2 Instrument support brackets for buoy anchors. From left to right: 45.7 cm long

ruler for scale, support that can hold two instruments, a variant of the first support model that can

also hold two instruments, and the latest support specially made for the SBE-56.

31

Figure A1.3 Thermograph support bracket on CCG buoy. The support bracket at the left is

welded under the hull behind the mooring chain buoy bracket, for measurements at 0.5 m depth.

The other larger support shown at the right is welded on the buoy foot at 2 m depth.

32

Figure A1.4 CCG buoy anchor. A thermograph support that can hold two instruments is bolted

to the top. Here the anchor height (L) is 0.9 m.

33

Figure A1.5 MLI buoy anchor. Three train wheels attached to a metal plate are used to anchor

buoys like the Viking MOB. The thermograph support bracket is bolted to the ring on top of the

anchor. The anchor height is 0.75 m. Anchors with only one or two train wheels are used for

lighter buoys.

34

Figure A1.6 MLI and EC buoys. MOB of first generation (left), Viking MOB (4.5 m high and

2.1 m wide, center), EC meteorological buoy usually deployed at Mont-Louis station (right).

35

APPENDIX 2. TABLES OF RECORD NUMBERS

Table A2.1 Number of records and gaps in water temperature time series starting in 1993. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gaps, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Bonne Bay (25) 11247 11247 11247 0 0 0 0 0

Borden (0.5) 14838 14838 14701 0 137 137 31 9

Ile Shag (10) 16201 16201 16201 0 0 0 0 0

Natashquan (12) 17496 17496 17496 0 0 0 0 0

Sept-Iles (2) 4881 4881 4881 0 0 0 0 0

Tadoussac (0.5) 4643 4643 4643 0 0 0 0 0

Table A2.2 Number of records and gaps in water temperature time series starting in 1994. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Bic (1) 8333 8333 8332 0 1 1 1 1

Bic (2) 8334 8334 8330 0 4 4 3 2

Bonne Bay (0.5) 7489 6984 6984 0 505 0 1 505

Bonne Bay (25) 7572 7572 7572 0 0 0 0 0

Borden (0.5) 11798 11798 11779 0 19 19 3 14

Grande-Rivière (2) 8458 8458 8457 0 1 1 1 1

Ile Shag (10) 12568 12568 12568 0 0 0 0 0

IML (0.5) 6374 6374 6374 0 0 0 0 0

IML PEM (12.4) 4985 4985 4985 0 0 0 0 0

La Romaine (1) 1667 1667 1667 0 0 0 0 0

La Romaine (2) 7375 7375 7375 0 0 0 0 0

Mont-Louis (0.5) 7525 7525 7525 0 0 0 0 0

Mont-Louis-exp (325) 7137 7137 7137 0 0 0 0 0

Natashquan (1) 7593 7593 7593 0 0 0 0 0

Port-Menier (2) 1459 1459 1459 0 0 0 0 0

Sept-Iles (2) 8737 8737 8736 0 1 1 1 1

Table A2.3 Number of records and gaps in water temperature time series starting in 1995. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Bic (1) 7658 7657 7657 0 0 0 0 0

Bic (2) 9253 9253 9253 0 0 0 0 0

Bonne Bay (0.5) 7373 7373 7373 0 0 0 0 0

Bonne Bay (25) 7792 7791 7791 0 0 0 0 0

Bonne Bay (25) 2099 2099 2099 0 0 0 0 0

Bonne Bay (25) 10069 10069 10069 0 0 0 0 0

Borden (0.5) 6765 6765 6765 0 0 0 0 0

36

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Borden (1.5) 6678 6678 6678 0 0 0 0 0

Borden (1.5) 16742 16742 16742 0 0 0 0 0

Grande-Rivière (2) 9149 9149 9149 0 0 0 0 0

Ile Shag (1) 6058 6058 6058 0 0 0 0 0

Ile Shag (10) 6056 6056 6056 0 0 0 0 0

Ile Shag (10) 11812 11812 11812 0 0 0 0 0

IML (0.5) 5391 5391 5391 0 0 0 0 0

IML (0.5) 3025 3025 3025 0 0 0 0 0

IML PEM (12.4) 15807 15807 15807 0 0 0 0 0

La Romaine (1) 6026 6025 6025 0 0 0 0 0

La Romaine (2) 8615 8615 8615 0 0 0 0 0

Mont-Louis (0.5) 10073 10073 10073 0 0 0 0 0

Mont-Louis (30) 7929 7928 7928 0 0 0 0 0

Natashquan (1) 7430 7430 7430 0 0 0 0 0

Port-Menier (2) 9397 9397 9397 0 0 0 0 0

Sept-Iles (2) 9303 9303 9303 0 0 0 0 0

Tadoussac (0.5) 7298 7298 7298 0 0 0 0 0

Table A2.4 Number of records and gaps in water temperature time series starting in 1996. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Bic (1) 7828 7828 7828 0 0 0 0 0

Bic (2) 7828 7828 7828 0 0 0 0 0

Bic (30.5) 7829 7829 7829 0 0 0 0 0

Bonne Bay (25) 6912 6912 6911 0 1 1 1 1

Borden (1.5) 12340 12340 12340 0 0 0 0 0

Grande-Rivière (2) 8299 8299 8299 0 0 0 0 0

Grande-Rivière (10) 8299 8299 8299 0 0 0 0 0

Havre-St-Pierre (2) 7680 7680 7680 0 0 0 0 0

Havre-St-Pierre (120) 7707 7707 7707 0 0 0 0 0

Ile Shag (1) 5908 5908 5908 0 0 0 0 0

Ile Shag (10) 5908 5908 5908 0 0 0 0 0

Ile Shag (10) 11176 11176 11176 0 0 0 0 0

IML (0.5) 4044 4044 4044 0 0 0 0 0

IML (0.5) 2354 2354 2354 0 0 0 0 0

IML PEM (12.4) 16397 16397 16397 0 0 0 0 0

La Perle (1) 7741 7741 7741 0 0 0 0 0

La Romaine (1) 8106 8106 8106 0 0 0 0 0

La Romaine (2) 8106 8106 8106 0 0 0 0 0

La Romaine (21.9) 8106 8106 8106 0 0 0 0 0

Mont-Louis (0.5) 361 361 361 0 0 0 0 0

Mont-Louis (0.5) 9178 9178 9178 0 0 0 0 0

Mont-Louis (30) 362 362 362 0 0 0 0 0

Mont-Louis (30) 7886 7885 7885 0 0 0 0 0

Natashquan (1) 7774 7774 7774 0 0 0 0 0

Natashquan (5.4) 7774 7774 7774 0 0 0 0 0

Port-Menier (2) 8602 8602 8602 0 0 0 0 0

Port-Menier (12.8) 8577 8577 8577 0 0 0 0 0

Riv.-au-Tonnerre (16) 7058 7058 7058 0 0 0 0 0

Sept-Iles (2) 8883 8883 8883 0 0 0 0 0

37

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Sept-Iles (21.9) 8883 8883 8883 0 0 0 0 0

Tadoussac (36.6) 16227 16227 16227 0 0 0 0 0

Table A2.5 Number of records and gaps in water temperature time series starting in 1997. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Bic (2) 7908 7908 7908 0 0 0 0 0

Bic (30.5) 7908 7908 7908 0 0 0 0 0

Bonne Bay (0.5) 8380 8380 8380 0 0 0 0 0

Bonne Bay (25) 8762 8762 8531 0 231 231 1 231

Bonne Bay (25) 8164 8164 8164 0 0 0 0 0

Borden (1.5) 14536 14536 14536 0 0 0 0 0

Grande-Rivière (2) 8200 8200 8200 0 0 0 0 0

Grande-Rivière (10) 8200 8200 8200 0 0 0 0 0

Havre-St-Pierre (1) 7150 7150 7150 0 0 0 0 0

Havre-St-Pierre (2) 6813 6813 6813 0 0 0 0 0

Havre-St-Pierre (120) 7148 7148 7148 0 0 0 0 0

Ile Shag (10) 6140 6140 6140 0 0 0 0 0

Ile Shag (10) 10615 10615 10615 0 0 0 0 0

IML (0.5) 1009 1009 1009 0 0 0 0 0

IML (0.5) 8389 8389 8389 0 0 0 0 0

IML PEM (12.4) 18429 18429 18426 0 3 3 1 3

La Perle (1) 7669 7669 7669 0 0 0 0 0

La Perle (26) 9360 9360 9360 0 0 0 0 0

La Romaine (2) 7626 7626 7626 0 0 0 0 0

Mont-Louis (0.5) 9310 9310 9310 0 0 0 0 0

Natashquan (5.4) 7340 7340 7340 0 0 0 0 0

Port-Menier (2) 8354 8354 8354 0 0 0 0 0

Port-Menier (12.8) 8355 8355 8355 0 0 0 0 0

Riv.-au-Tonnerre (16) 8425 8425 8425 0 0 0 0 0

Sept-Iles (2) 9098 9098 9098 0 0 0 0 0

Sept-Iles (21.9) 9100 9100 9100 0 0 0 0 0

Table A2.6 Number of records and gaps in water temperature time series starting in 1998. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Banc Beaugé (1) 7870 7870 7870 0 0 0 0 0

Banc Beaugé (97) 7866 7866 7866 0 0 0 0 0

Bic (2) 7958 7958 7958 0 0 0 0 0

Bic (30.5) 7956 7956 7956 0 0 0 0 0

Bonne Bay (0.5) 6988 6988 6987 0 1 1 1 1

Bonne Bay (25) 6988 6988 6988 0 0 0 0 0

Bonne Bay (25) 15939 15939 15939 0 0 0 0 0

Borden (1.5) 17502 17502 17502 0 0 0 0 0

Grande-Rivière (2) 10407 10407 10407 0 0 0 0 0

38

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Grande-Rivière (10) 10407 10407 10407 0 0 0 0 0

Havre-St-Pierre (1) 8264 8264 8264 0 0 0 0 0

Havre-St-Pierre (120) 8261 8261 8261 0 0 0 0 0

IML (0.5) 6867 6867 6867 0 0 0 0 0

IML PEM (12.4) 16739 16739 16739 0 0 0 0 0

Ile Shag (10) 7386 7386 7386 0 0 0 0 0

Ile Shag (10) 10900 10900 10900 0 0 0 0 0

Irving Whale (0.5) 10836 10836 10836 0 0 0 0 0

Irving Whale (1) 10836 10836 10836 0 0 0 0 0

Irving Whale (67) 7960 7959 7959 0 0 0 0 0

La Perle (1) 7621 7621 7621 0 0 0 0 0

La Perle (26) 7480 7479 7479 0 0 0 0 0

La Romaine (2) 8053 8053 8053 0 0 0 0 0

La Romaine (21.9) 8054 8054 8054 0 0 0 0 0

Mont-Louis (0.35) 9562 9562 9562 0 0 0 0 0

Mont-Louis (0.5) 9562 9562 9562 0 0 0 0 0

Mont-Louis (0.65) 9562 9562 9562 0 0 0 0 0

Mont-Louis (0.90) 9562 9562 9562 0 0 0 0 0

Mont-Louis (1.1) 9562 9562 9562 0 0 0 0 0

Mont-Louis (2.1) 9562 9562 9562 0 0 0 0 0

Mont-Louis (30) 9562 9562 9562 0 0 0 0 0

Natashquan (1) 8153 8153 8153 0 0 0 0 0

Natashquan (5.4) 8154 8154 8154 0 0 0 0 0

Port-Menier (2) 6385 6385 6385 0 0 0 0 0

Riv.-au-Tonnerre (1) 8293 8293 8293 0 0 0 0 0

Riv.-au-Tonnerre (16) 1263 1263 1263 0 0 0 0 0

Riv.-au-Tonnerre (16) 1263 1263 1263 0 0 0 0 0

Sept-Iles (2) 8685 8685 8685 0 0 0 0 0

Tadoussac (0.5) 9592 9592 9592 0 0 0 0 0

Tadoussac (36.6) 7898 7897 7897 0 0 0 0 0

Tadoussac (36.6) 1550 1550 1550 0 0 0 0 0

Table A2.7 Number of records and gaps in water temperature time series starting in 1999. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 10610 10610 10610 0 0 0 0 0

Banc Beaugé (1) 8187 8187 8187 0 0 0 0 0

Banc Beaugé (97) 8187 8187 8187 0 0 0 0 0

Bic (1) 8695 8695 8695 0 0 0 0 0

Bic (2) 8695 8695 8695 0 0 0 0 0

Bic (30.5) 8695 8695 8695 0 0 0 0 0

Blanc-Sablon (1) 8386 8386 8386 0 0 0 0 0

Blanc-Sablon (22) 8386 8386 8386 0 0 0 0 0

Bonne Bay (0.5) 6580 6579 6579 0 0 0 0 0

Bonne Bay (25) 7154 7154 7154 0 0 0 0 0

Bonne Bay (25) 9893 9893 9889 0 4 4 2 2

Borden (1.5) 11720 11720 11720 0 0 0 0 0

Borden (1.5) 7097 7097 7097 0 0 0 0 0

Grande-Rivière (2) 10225 10225 10225 0 0 0 0 0

Grande-Rivière (10) 10225 10225 10225 0 0 0 0 0

39

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Havre-St-Pierre (1) 9048 9048 9048 0 0 0 0 0

Havre-St-Pierre (120) 9048 9048 9048 0 0 0 0 0

IML (0.5) 28317 28311 28310 1 7 0 2 6

IML (0.5) 28317 28311 28310 1 7 0 2 6

IML PEM (12.4) 9261 9261 9261 0 0 0 0 0

Ile Shag (1) 6466 6466 6466 0 0 0 0 0

Ile Shag (10) 5891 5891 5891 0 0 0 0 0

Ile Shag (10) 11048 11044 11044 0 4 0 2 2

Ile Shag-exp (0.2) 6466 6466 6466 0 0 0 0 0

Irving Whale (0.5) 9351 9351 9351 0 0 0 0 0

Irving Whale (1) 9351 9351 9351 0 0 0 0 0

La Perle (1) 7920 7920 7920 0 0 0 0 0

La Perle (26) 9835 9835 9835 0 0 0 0 0

La Romaine (2) 8846 8846 8846 0 0 0 0 0

La Romaine (21.9) 8846 8846 8846 0 0 0 0 0

Mont-Louis (0.35) 10125 10125 10074 0 51 51 1 51

Mont-Louis (0.5) 10125 10125 10074 0 51 51 1 51

Mont-Louis (1.1) 10125 10125 10074 0 51 51 1 51

Mont-Louis (2.1) 10125 10125 10074 0 51 51 1 51

Mont-Louis (30) 10125 10125 10024 0 101 101 7 86

Natashquan (1) 8973 8973 8973 0 0 0 0 0

Natashquan (5.4) 8973 8973 8973 0 0 0 0 0

Port-Menier (2) 10205 10205 10205 0 0 0 0 0

Port-Menier (12.8) 10205 10205 10205 0 0 0 0 0

Riv.-au-Tonnerre (16) 2144 2144 2144 0 0 0 0 0

Riv.-au-Tonnerre (16) 2144 2144 2144 0 0 0 0 0

Sept-Iles (2) 10686 10686 10686 0 0 0 0 0

Sept-Iles (21.9) 10686 10686 10686 0 0 0 0 0

Tadoussac (0.5) 9690 9690 9209 0 481 481 1 481

Table A2.8 Number of records and gaps in water temperature time series starting in 2000. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 9144 9144 9144 0 0 0 0 0

Baie-Comeau (82.3) 9144 9144 9126 0 18 18 1 18

Banc Beaugé (1) 7881 7881 7881 0 0 0 0 0

Bic (2) 8476 8476 8476 0 0 0 0 0

Bic (30.5) 8476 8476 8476 0 0 0 0 0

Blanc-Sablon (1) 7593 7593 7593 0 0 0 0 0

Blanc-Sablon (22) 7593 7593 7593 0 0 0 0 0

Bonne Bay (0.5) 7721 7721 7721 0 0 0 0 0

Bonne Bay (25) 7721 7721 7721 0 0 0 0 0

Bonne Bay (25) 9120 9120 9120 0 0 0 0 0

Borden (1.5) 28189 28189 28189 0 0 0 0 0

Grande-Rivière (2) 8944 8944 8944 0 0 0 0 0

Grande-Rivière (10) 8944 8944 8944 0 0 0 0 0

Havre-St-Pierre (1) 7814 7814 7814 0 0 0 0 0

Havre-St-Pierre (120) 7752 7752 7752 0 0 0 0 0

IML (0.5) 9066 9066 9066 0 0 0 0 0

IML PEM (12.4) 12188 12188 12188 0 0 0 0 0

40

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Ile Shag (1) 6384 6384 6068 0 316 316 1 316

Ile Shag (10) 6384 6384 6384 0 0 0 0 0

Ile Shag (10) 11467 11467 11467 0 0 0 0 0

Ile Shag-exp (0.2) 6384 6384 6068 0 316 316 1 316

Irving Whale (0.5) 7484 7484 7484 0 0 0 0 0

Irving Whale (1) 7460 7460 7460 0 0 0 0 0

La Perle (1) 9936 9936 9936 0 0 0 0 0

La Perle (26) 9936 9936 9936 0 0 0 0 0

La Romaine (1) 7782 7782 7782 0 0 0 0 0

La Romaine (2) 7782 7782 7782 0 0 0 0 0

La Romaine (21.9) 7781 7781 7781 0 0 0 0 0

Mont-Louis (0.35) 8978 8978 8978 0 0 0 0 0

Mont-Louis (0.5) 8979 8979 8979 0 0 0 0 0

Mont-Louis (1.1) 8978 8978 8978 0 0 0 0 0

Mont-Louis (2.1) 8978 8978 8978 0 0 0 0 0

Mont-Louis (30) 8978 8978 8978 0 0 0 0 0

Natashquan (1) 7808 7808 7808 0 0 0 0 0

Natashquan (5.4) 7808 7808 7808 0 0 0 0 0

Port-Menier (2) 8353 8353 8353 0 0 0 0 0

Port-Menier (12.8) 8354 8354 8354 0 0 0 0 0

Riv.-au-Tonnerre (1) 7788 7788 7788 0 0 0 0 0

Riv.-au-Tonnerre (16) 667 667 667 0 0 0 0 0

Riv.-au-Tonnerre (16) 667 667 667 0 0 0 0 0

Sept-Iles (2) 8730 8730 8730 0 0 0 0 0

Sept-Iles (21.9) 8730 8730 8730 0 0 0 0 0

Tadoussac (0.5) 9489 9489 9463 0 26 26 2 24

Tadoussac (36.6) 9489 9489 9465 0 24 24 2 22

Table A2.9 Number of records and gaps in water temperature time series starting in 2001. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 9062 9062 9062 0 0 0 0 0

Banc Beaugé (1) 8975 8975 8975 0 0 0 0 0

Banc Beaugé (97) 8975 8975 8975 0 0 0 0 0

Bic (1) 7724 7724 7724 0 0 0 0 0

Bic (2) 7723 7723 7723 0 0 0 0 0

Bic (30.5) 7724 7724 7724 0 0 0 0 0

Blanc-Sablon (1) 7388 7388 7388 0 0 0 0 0

Blanc-Sablon (22) 7388 7388 7388 0 0 0 0 0

Bonne Bay (0.5) 7823 7823 7823 0 0 0 0 0

Bonne Bay (25) 7849 7849 7849 0 0 0 0 0

Bonne Bay (25) 8838 8838 8838 0 0 0 0 0

Grande-Rivière (2) 8686 8686 8686 0 0 0 0 0

Grande-Rivière (10) 8686 8686 8686 0 0 0 0 0

Havre-St-Pierre (1) 9255 9255 9255 0 0 0 0 0

IML (0.5) 7789 7789 7789 0 0 0 0 0

IML PEM (12.4) 22415 22414 22414 0 0 0 0 0

Ile Shag (1) 6762 6762 6762 0 0 0 0 0

Ile Shag (10) 6761 6761 6761 0 0 0 0 0

Ile Shag (10) 9275 9275 9275 0 0 0 0 0

41

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Ile Shag-exp (0.2) 6762 6762 6762 0 0 0 0 0

Irving Whale (0.5) 9851 9851 9851 0 0 0 0 0

Irving Whale (1) 9851 9851 9851 0 0 0 0 0

Irving Whale (67) 9851 9851 9851 0 0 0 0 0

La Perle (1) 8154 8154 8154 0 0 0 0 0

La Perle (26) 8155 8155 8155 0 0 0 0 0

La Romaine (1) 8974 8974 8974 0 0 0 0 0

La Romaine (2) 8973 8973 8973 0 0 0 0 0

La Romaine (21.9) 8974 8974 8974 0 0 0 0 0

Mont-Louis (0.35) 9394 9394 9394 0 0 0 0 0

Mont-Louis (0.5) 9394 9394 9394 0 0 0 0 0

Mont-Louis (1.1) 9394 9394 9394 0 0 0 0 0

Mont-Louis (2) 9394 9392 9392 0 2 0 2 1

Mont-Louis (2.1) 9394 9394 9394 0 0 0 0 0

Mont-Louis (30) 9394 9394 9394 0 0 0 0 0

Natashquan (1) 9251 9251 9251 0 0 0 0 0

Natashquan (5.4) 9251 9251 9251 0 0 0 0 0

Port-Menier (2) 8839 8839 8839 0 0 0 0 0

Port-Menier (12.8) 8839 8839 8839 0 0 0 0 0

Riv.-au-Tonnerre (1) 9332 9332 9332 0 0 0 0 0

Riv.-au-Tonnerre (16) 8851 8851 8851 0 0 0 0 0

Sept-Iles (1) 8966 8966 8966 0 0 0 0 0

Sept-Iles (21.9) 8966 8966 8966 0 0 0 0 0

Tadoussac (0.5) 9404 9404 9394 0 10 10 1 10

Table A2.10 Number of records and gaps in water temperature time series starting in 2002. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 9560 9560 9560 0 0 0 0 0

Baie-Comeau (82.3) 1361 1361 1361 0 0 0 0 0

Baie-Comeau (82.3) 7222 7222 7222 0 0 0 0 0

Banc Beaugé (1) 7213 7213 7213 0 0 0 0 0

Banc Beaugé (97) 7213 7213 7213 0 0 0 0 0

Bic (1) 8437 8437 8437 0 0 0 0 0

Bic (2) 8437 8437 8437 0 0 0 0 0

Bic (30.5) 8437 8437 8437 0 0 0 0 0

Blanc-Sablon (1) 7017 7017 7017 0 0 0 0 0

Blanc-Sablon (22) 7016 7016 7016 0 0 0 0 0

Bonne Bay (0.5) 4444 4444 4444 0 0 0 0 0

Bonne Bay (25) 8142 8142 8119 0 23 23 1 23

Bonne Bay (25) 10620 10620 10620 0 0 0 0 0

Borden (1.5) 26894 26893 26893 0 0 0 0 0

Grande-Rivière (2) 8884 8884 8884 0 0 0 0 0

Grande-Rivière (10) 8884 8884 8884 0 0 0 0 0

Havre-St-Pierre (1) 7528 7528 7528 0 0 0 0 0

Havre-St-Pierre (120) 7528 7528 7528 0 0 0 0 0

IML (0.5) 8527 8527 8527 0 0 0 0 0

IML PEM (12.4) 26466 26466 26466 0 0 0 0 0

Ile Shag (1) 7095 7095 7095 0 0 0 0 0

Ile Shag (10) 7094 7094 7094 0 0 0 0 0

42

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Ile Shag (10) 10473 10473 10473 0 0 0 0 0

Ile Shag-exp (0.2) 7095 7095 7095 0 0 0 0 0

Irving Whale (0.5) 10114 10114 10114 0 0 0 0 0

Irving Whale (1) 10114 10114 10114 0 0 0 0 0

Irving Whale (67) 10114 10114 10114 0 0 0 0 0

La Perle (1) 10556 10556 10556 0 0 0 0 0

La Perle (26) 10556 10556 10556 0 0 0 0 0

La Romaine (1) 7291 7291 7291 0 0 0 0 0

La Romaine (2) 7291 7291 7291 0 0 0 0 0

La Romaine (21.9) 7291 7291 7291 0 0 0 0 0

La Tabatière (1) 6600 6600 6600 0 0 0 0 0

La Tabatière (39) 7170 7170 7170 0 0 0 0 0

Mont-Louis (0.35) 10601 10601 10601 0 0 0 0 0

Mont-Louis (0.5) 10601 10601 10601 0 0 0 0 0

Mont-Louis (1.1) 10601 10601 10601 0 0 0 0 0

Mont-Louis (2.1) 10601 10601 10601 0 0 0 0 0

Mont-Louis (30) 10600 10600 10600 0 0 0 0 0

Natashquan (1) 7552 7552 7552 0 0 0 0 0

Natashquan (5.4) 7553 7553 7553 0 0 0 0 0

Port-Menier (2) 9023 9023 9023 0 0 0 0 0

Port-Menier (12.8) 9023 9023 9023 0 0 0 0 0

Rimouski (0.5) 11521 11511 11510 0 11 1 6 3

Riv.-au-Tonnerre (1) 7639 7639 7639 0 0 0 0 0

Riv.-au-Tonnerre (16) 7639 7639 7639 0 0 0 0 0

Sept-Iles (1) 9460 9460 9460 0 0 0 0 0

Sept-Iles (21.9) 9460 9460 9460 0 0 0 0 0

Tadoussac (0.5) 10032 10032 10032 0 0 0 0 0

Tadoussac (36.6) 10032 10032 10032 0 0 0 0 0

Tadoussac-exp (43) 20333 20318 20318 0 15 0 15 1

Table A2.11 Number of records and gaps in water temperature time series starting in 2003. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 8741 8741 8741 0 0 0 0 0

Baie-Comeau (82.3) 8738 8738 8738 0 0 0 0 0

Baie-Comeau (82.3) 8431 8431 8431 0 0 0 0 0

Banc Beaugé (1) 5938 5938 5938 0 0 0 0 0

Banc Beaugé (97) 5939 5939 5939 0 0 0 0 0

Bic (1) 7374 7374 7374 0 0 0 0 0

Bic (2) 7374 7374 7374 0 0 0 0 0

Bic (30.5) 7374 7374 7374 0 0 0 0 0

Blanc-Sablon (1) 5805 5805 5805 0 0 0 0 0

Blanc-Sablon (22) 16573 16573 16573 0 0 0 0 0

Bonne Bay (25) 8355 8355 8355 0 0 0 0 0

Grande-Rivière (2) 8554 8554 8554 0 0 0 0 0

Grande-Rivière (10) 8554 8554 8554 0 0 0 0 0

Havre-St-Pierre (1) 6275 6275 6275 0 0 0 0 0

Havre-St-Pierre (120) 6275 6275 6275 0 0 0 0 0

IML (0.5) 7428 7428 7428 0 0 0 0 0

Ile Shag (1) 6005 6005 6005 0 0 0 0 0

43

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Ile Shag (10) 5765 5765 5765 0 0 0 0 0

Ile Shag (10) 10417 10417 10417 0 0 0 0 0

Ile Shag-exp (0.2) 6005 6005 6005 0 0 0 0 0

Irving Whale (0.5) 9408 9404 9404 0 4 0 4 1

Irving Whale (1) 9408 9408 9408 0 0 0 0 0

Irving Whale (67) 9409 9409 9409 0 0 0 0 0

La Perle (1) 9416 9416 9416 0 0 0 0 0

La Perle (26) 9416 9416 9416 0 0 0 0 0

La Romaine (1) 5938 5938 5938 0 0 0 0 0

La Romaine (2) 5938 5938 5938 0 0 0 0 0

La Romaine (21.9) 5938 5938 5938 0 0 0 0 0

La Tabatière (1) 5800 5800 5800 0 0 0 0 0

La Tabatière (39) 5799 5799 5799 0 0 0 0 0

Mont-Louis (0.35) 9688 9688 9688 0 0 0 0 0

Mont-Louis (0.5) 9689 9689 9689 0 0 0 0 0

Mont-Louis (1.1) 9688 9688 9688 0 0 0 0 0

Mont-Louis (2.1) 9660 9660 9660 0 0 0 0 0

Mont-Louis (30) 9642 9642 9642 0 0 0 0 0

Natashquan (1) 6048 6048 6048 0 0 0 0 0

Natashquan (5.4) 6048 6048 6048 0 0 0 0 0

Rimouski (0.5) 14790 14790 14790 0 0 0 0 0

Riv.-au-Tonnerre (1) 6345 6345 6345 0 0 0 0 0

Riv.-au-Tonnerre (16) 6345 6345 6345 0 0 0 0 0

Sept-Iles (1) 8888 8888 8888 0 0 0 0 0

Sept-Iles (21.9) 8888 8888 8888 0 0 0 0 0

Tadoussac (0.5) 8749 8749 8749 0 0 0 0 0

Table A2.12 Number of records and gaps in water temperature time series starting in 2004. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 8575 8575 8575 0 0 0 0 0

Baie-Comeau (82.3) 8575 8575 8575 0 0 0 0 0

Baie-Comeau (82.3) 8566 8566 8566 0 0 0 0 0

Banc Beaugé (1) 7001 7001 7001 0 0 0 0 0

Banc Beaugé (97) 7001 7001 7001 0 0 0 0 0

Bic (1) 8426 8426 8426 0 0 0 0 0

Bic (2) 8426 8426 8426 0 0 0 0 0

Bic (30.5) 8426 8426 8426 0 0 0 0 0

Blanc-Sablon (1) 6898 6898 6898 0 0 0 0 0

Blanc-Sablon (22) 6897 6897 6897 0 0 0 0 0

Bonne Bay (0.5) 7694 7694 7694 0 0 0 0 0

Bonne Bay (25) 7694 7694 7694 0 0 0 0 0

Grande-Rivière (2) 7546 7546 7546 0 0 0 0 0

Grande-Rivière (10) 7546 7546 7546 0 0 0 0 0

Havre-St-Pierre (0.5) 7191 7191 7191 0 0 0 0 0

Havre-St-Pierre (1) 7191 7191 7191 0 0 0 0 0

IML (0.5) 8444 8444 8444 0 0 0 0 0

IML PEM (12.4) 6823 6823 6823 0 0 0 0 0

IML PEM (12.4) 16227 16227 16227 0 0 0 0 0

Ile Shag (0.2) 7026 7026 7026 0 0 0 0 0

44

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Ile Shag (1) 7026 7026 7026 0 0 0 0 0

Ile Shag (10) 7026 7026 7026 0 0 0 0 0

Ile Shag (10) 10025 10025 10025 0 0 0 0 0

Irving Whale (0.5) 8988 8988 8988 0 0 0 0 0

Irving Whale (1) 8988 8988 8988 0 0 0 0 0

Irving Whale (67) 8987 8987 8987 0 0 0 0 0

La Perle (1) 9252 9252 8843 0 409 409 1 409

La Romaine (2) 6885 6885 6885 0 0 0 0 0

La Romaine (21.9) 6885 6885 6885 0 0 0 0 0

La Tabatière (1) 6891 6891 6891 0 0 0 0 0

La Tabatière (39) 6891 6891 6891 0 0 0 0 0

Mont-Louis (0.5) 7949 7949 7949 0 0 0 0 0

Mont-Louis (1.1) 7949 7949 7949 0 0 0 0 0

Mont-Louis (2.1) 7950 7950 7950 0 0 0 0 0

Mont-Louis (30) 7949 7949 7949 0 0 0 0 0

Natashquan (1) 6902 6902 6902 0 0 0 0 0

Natashquan (5.4) 6902 6902 6902 0 0 0 0 0

Old Belle Isle (105) 14444 14436 14436 0 8 0 8 1

Old Belle Isle (105) 18108 18108 18108 0 0 0 0 0

Port-Menier (2) 7848 7848 7848 0 0 0 0 0

Rimouski (0.5) 16890 15875 15875 0 1015 0 376 102

Riv.-au-Tonnerre (1) 7247 7247 7247 0 0 0 0 0

Riv.-au-Tonnerre (16) 7247 7247 7229 0 18 18 1 18

Sept-Iles (1) 8447 8447 8447 0 0 0 0 0

Sept-Iles (2) 8447 8447 8447 0 0 0 0 0

Sept-Iles (21.9) 8447 8447 8447 0 0 0 0 0

Shediac Valley (0.5) 15456 14561 14561 0 895 0 202 102

Tadoussac (0.5) 9405 9405 9405 0 0 0 0 0

Tadoussac (36.6) 9406 9406 9406 0 0 0 0 0

Table A2.13 Number of records and gaps in water temperature time series starting in 2005. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 9360 9360 9360 0 0 0 0 0

Baie-Comeau (82.3) 9674 9674 9674 0 0 0 0 0

Banc Beaugé (0.5) 16533 16504 16497 0 36 7 4 23

Banc Beaugé (1) 8266 8266 8266 0 0 0 0 0

Banc Beaugé (97) 8266 8266 8266 0 0 0 0 0

Bic (1) 8031 8031 8031 0 0 0 0 0

Bic (2) 8054 8054 8054 0 0 0 0 0

Bic (30.5) 8054 8054 8054 0 0 0 0 0

Blanc-Sablon (1) 8215 8215 8215 0 0 0 0 0

Blanc-Sablon (22) 8215 8215 8215 0 0 0 0 0

Courant de Gaspé (0.5) 1424 1424 1424 0 0 0 0 0

Courant de Gaspé (165) 711 711 711 0 0 0 0 0

Grande-Rivière (2) 9595 9595 9595 0 0 0 0 0

Grande-Rivière (10) 9594 9594 9594 0 0 0 0 0

Gyre Anticosti (0.5) 13564 13562 13562 0 2 0 2 1

Gyre Anticosti (337) 6781 6781 6781 0 0 0 0 0

Havre-St-Pierre (0.5) 8253 8253 8253 0 0 0 0 0

45

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Havre-St-Pierre (1) 8253 8253 8253 0 0 0 0 0

Havre-St-Pierre (120) 8252 8252 8252 0 0 0 0 0

IML (0.5) 8021 8021 8021 0 0 0 0 0

IML PEM (12.4) 2065 2065 2065 0 0 0 0 0

IML PEM (12.4) 19686 19686 19686 0 0 0 0 0

Ile Shag (10) 6984 6984 6984 0 0 0 0 0

Ile Shag (10) 10174 10174 10174 0 0 0 0 0

Irving Whale (0.5) 8247 8247 8247 0 0 0 0 0

Irving Whale (1) 8738 8738 8738 0 0 0 0 0

Irving Whale (67) 8738 8738 8738 0 0 0 0 0

La Perle (1) 8674 8674 8674 0 0 0 0 0

La Perle (26) 8674 8674 8674 0 0 0 0 0

La Romaine (1) 8551 8551 8551 0 0 0 0 0

La Romaine (2) 8551 8551 8551 0 0 0 0 0

La Romaine (21.9) 8551 8551 8551 0 0 0 0 0

La Tabatière (1) 8166 8166 8166 0 0 0 0 0

La Tabatière (39) 8166 8166 8166 0 0 0 0 0

Mont-Louis (0.35) 9814 9814 9814 0 0 0 0 0

Mont-Louis (0.5) 9814 9814 9814 0 0 0 0 0

Mont-Louis (1.1) 9814 9814 9814 0 0 0 0 0

Mont-Louis (2.1) 9814 9814 9814 0 0 0 0 0

Mont-Louis (30) 9814 9814 9814 0 0 0 0 0

Natashquan (1) 8265 8265 8265 0 0 0 0 0

Natashquan (5.4) 8265 8265 8265 0 0 0 0 0

Old Belle Isle (105) 10503 10503 10503 0 0 0 0 0

Port-Menier (2) 9276 9276 9276 0 0 0 0 0

Port-Menier (12.8) 9275 9275 9275 0 0 0 0 0

Rimouski (0.5) 16061 15678 15678 0 383 0 44 133

Rimouski (330) 8031 8031 8031 0 0 0 0 0

Riv.-au-Tonnerre (1) 8230 8230 8230 0 0 0 0 0

Riv.-au-Tonnerre (16) 8230 8230 8230 0 0 0 0 0

Sept-Iles (1) 9603 9603 9603 0 0 0 0 0

Sept-Iles (2) 9603 9603 9603 0 0 0 0 0

Sept-Iles (21.9) 9603 9603 9603 0 0 0 0 0

Shediac Valley (0.5) 19161 19160 19160 0 1 0 1 1

Shediac Valley (85.8) 9581 9581 9581 0 0 0 0 0

Tadoussac (0.5) 10598 10598 10598 0 0 0 0 0

Tadoussac (1) 10598 10598 10598 0 0 0 0 0

Table A2.14 Number of records and gaps in water temperature time series starting in 2006. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 8543 8543 8543 0 0 0 0 0

Baie-Comeau (82.3) 8543 8543 8543 0 0 0 0 0

Baie-Comeau (82.3) 8145 8145 8145 0 0 0 0 0

Banc Beaugé (0.5) 16747 15299 15298 1 1449 0 1 1449

Banc Beaugé (1) 8373 8373 8373 0 0 0 0 0

Banc Beaugé (97) 8373 8373 8373 0 0 0 0 0

Bic (1) 8063 8063 8063 0 0 0 0 0

Bic (2) 8063 8063 8063 0 0 0 0 0

46

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Bic (30.5) 8063 8063 8063 0 0 0 0 0

Blanc-Sablon (1) 7537 7537 7537 0 0 0 0 0

Blanc-Sablon (22) 7536 7536 7536 0 0 0 0 0

Churchill (0.5) 3940 3940 3940 0 0 0 0 0

Courant de Gaspé (0.5) 16264 16201 16201 0 63 0 16 24

Grande-Rivière (2) 9136 9136 9136 0 0 0 0 0

Grande-Rivière (10) 9136 9136 9136 0 0 0 0 0

Gyre Anticosti (0.5) 16238 16238 16238 0 0 0 0 0

Gyre Anticosti (337) 8119 8119 8119 0 0 0 0 0

Havre-St-Pierre (0.5) 7494 7494 7494 0 0 0 0 0

Havre-St-Pierre (1) 7494 7494 7494 0 0 0 0 0

Havre-St-Pierre (120) 7493 7493 7493 0 0 0 0 0

IML PEM (12.4) 7870 7870 7870 0 0 0 0 0

Ile Shag (10) 7249 7249 7249 0 0 0 0 0

Ile Shag (10) 12196 12196 12196 0 0 0 0 0

Irving Whale (0.5) 9056 9056 9056 0 0 0 0 0

Irving Whale (1) 9056 9056 9056 0 0 0 0 0

Irving Whale (67) 9056 9056 9056 0 0 0 0 0

La Perle (1) 572 572 572 0 0 0 0 0

La Perle (26) 571 571 571 0 0 0 0 0

La Romaine (1) 7539 7539 7539 0 0 0 0 0

La Romaine (2) 7554 7554 7554 0 0 0 0 0

La Romaine (21.9) 7539 7539 7539 0 0 0 0 0

La Tabatière (1) 7669 7669 7669 0 0 0 0 0

La Tabatière (39) 7669 7669 7669 0 0 0 0 0

Mont-Louis (0.35) 8892 8892 8892 0 0 0 0 0

Mont-Louis (0.5) 8893 8893 8893 0 0 0 0 0

Mont-Louis (1.1) 8893 8893 8893 0 0 0 0 0

Mont-Louis (2.1) 8893 8893 8893 0 0 0 0 0

Mont-Louis (30) 8892 8892 8892 0 0 0 0 0

Natashquan (1) 7535 7535 7535 0 0 0 0 0

Natashquan (5.4) 7535 7535 7535 0 0 0 0 0

Old Belle Isle (105) 6613 6613 6613 0 0 0 0 0

Old Belle Isle (105) 17476 17476 17476 0 0 0 0 0

Port-Menier (2) 8641 8641 8641 0 0 0 0 0

Port-Menier (12.8) 8641 8641 8641 0 0 0 0 0

Rimouski (0.5) 15929 15927 15927 0 2 0 1 2

Riv.-au-Tonnerre (1) 7494 7494 7494 0 0 0 0 0

Riv.-au-Tonnerre (16) 7494 7494 7494 0 0 0 0 0

Sept-Iles (1) 8775 8775 8775 0 0 0 0 0

Sept-Iles (21.9) 8775 8775 8775 0 0 0 0 0

Shediac Valley (0.5) 16988 13675 13672 3 3316 0 95 381

Shediac Valley (85.8) 8552 8552 8552 0 0 0 0 0

Tadoussac (0.5) 10374 10374 10374 0 0 0 0 0

Tadoussac (1) 10374 10374 10374 0 0 0 0 0

Tadoussac (36.6) 10374 10374 10374 0 0 0 0 0

47

Table A2.15 Number of records and gaps in water temperature time series starting in 2007. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 9505 9505 9505 0 0 0 0 0

Baie-Comeau (82.3) 18209 18209 18209 0 0 0 0 0

Banc Beaugé (0.5) 15817 15684 15684 0 133 0 3 50

Banc Beaugé (1) 8197 8197 8197 0 0 0 0 0

Banc Beaugé (97) 8197 8197 8197 0 0 0 0 0

Belle Isle (71) 6944 6944 6944 0 0 0 0 0

Belle Isle (71) 17674 17674 17674 0 0 0 0 0

Bic (1) 8498 8498 8498 0 0 0 0 0

Bic (2) 8498 8498 8498 0 0 0 0 0

Bic (30.5) 7034 7034 7034 0 0 0 0 0

Blanc-Sablon (1) 7202 7202 7202 0 0 0 0 0

Churchill (0.5) 3583 3583 3583 0 0 0 0 0

Courant de Gaspé (0.5) 14041 14041 14041 0 0 0 0 0

Courant de Gaspé (165) 7021 7021 7021 0 0 0 0 0

Grande-Rivière (2) 9025 9025 9025 0 0 0 0 0

Grande-Rivière (10) 9025 9025 9025 0 0 0 0 0

Gyre Anticosti (0.5) 18753 18753 18753 0 0 0 0 0

Gyre Anticosti (337) 9376 9376 9376 0 0 0 0 0

Havre-St-Pierre (0.5) 8559 8559 8559 0 0 0 0 0

Havre-St-Pierre (1) 8560 8560 8560 0 0 0 0 0

Havre-St-Pierre (120) 8560 8560 8560 0 0 0 0 0

IML (0.5) 8502 8502 8502 0 0 0 0 0

IML PEM (12.4) 8730 8729 8729 0 0 0 0 0

IML PEM (12.4) 19282 19282 19282 0 0 0 0 0

Ile Shag (1) 4459 4459 4459 0 0 0 0 0

Ile Shag (1) 4459 4459 4459 0 0 0 0 0

Ile Shag (10) 5322 5322 5322 0 0 0 0 0

Ile Shag (10) 11765 11765 11765 0 0 0 0 0

Irving Whale (0.5) 11042 11042 11042 0 0 0 0 0

Irving Whale (1) 9634 9634 9634 0 0 0 0 0

Irving Whale (67) 9634 9634 9634 0 0 0 0 0

La Perle (1) 9225 9225 9225 0 0 0 0 0

La Perle (26) 10611 10611 10611 0 0 0 0 0

La Romaine (1) 8378 8378 8378 0 0 0 0 0

La Romaine (2) 8378 8378 8378 0 0 0 0 0

La Romaine (21.9) 8378 8378 8378 0 0 0 0 0

La Tabatière (1) 7941 7941 7941 0 0 0 0 0

La Tabatière (39) 7941 7941 7941 0 0 0 0 0

Mont-Louis (0.35) 9475 9475 9475 0 0 0 0 0

Mont-Louis (0.5) 9475 9475 9475 0 0 0 0 0

Mont-Louis (1.1) 9475 9475 9475 0 0 0 0 0

Mont-Louis (2.1) 9475 9475 9475 0 0 0 0 0

Mont-Louis (30) 5246 5246 5246 0 0 0 0 0

Natashquan (1) 8426 8426 8426 0 0 0 0 0

Natashquan (5.4) 8426 8426 8426 0 0 0 0 0

Port-Menier (2) 9133 9133 9133 0 0 0 0 0

Port-Menier (12.8) 9133 9133 9133 0 0 0 0 0

Rimouski (0.5) 20175 20173 20173 0 2 0 1 2

Rimouski (330) 10088 10088 10088 0 0 0 0 0

48

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Riv.-au-Tonnerre (1) 8562 8562 8562 0 0 0 0 0

Riv.-au-Tonnerre (16) 8561 8561 8561 0 0 0 0 0

Sept-Iles (1) 9505 9505 9505 0 0 0 0 0

Sept-Iles (2) 9505 9505 9505 0 0 0 0 0

Sept-Iles (21.9) 9505 9505 9505 0 0 0 0 0

Shediac Valley (0.5) 18228 15092 15092 0 3136 0 50 278

Shediac Valley (85.8) 9120 9120 9120 0 0 0 0 0

Tadoussac (1) 10307 10307 10305 0 2 2 1 2

Tadoussac (36.6) 10307 10307 10305 0 2 2 1 2

Table A2.16 Number of records and gaps in water temperature time series starting in 2008. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 8689 8689 8689 0 0 0 0 0

Baie-Comeau (82.3) 17037 17037 17037 0 0 0 0 0

Banc Beaugé (0.5) 14962 14960 14955 5 7 0 7 1

Banc Beaugé (1) 7481 7481 7481 0 0 0 0 0

Banc Beaugé (97) 7481 7481 7481 0 0 0 0 0

Belle Isle (71) 10137 10137 10137 0 0 0 0 0

Bic (1) 8154 8154 8154 0 0 0 0 0

Bic (2) 8154 8154 8154 0 0 0 0 0

Bic (30.5) 8154 8154 8154 0 0 0 0 0

Blanc-Sablon (1) 6519 6519 6519 0 0 0 0 0

Blanc-Sablon (22) 6519 6519 6519 0 0 0 0 0

Borden (1.25) 20993 20993 20993 0 0 0 0 0

Borden (1.25) 60 60 60 0 0 0 0 0

Courant de Gaspé (0.5) 16654 16653 16653 0 1 0 1 1

Courant de Gaspé (165) 8328 8328 8328 0 0 0 0 0

Grande-Rivière (2) 8318 8318 8318 0 0 0 0 0

Grande-Rivière (10) 8318 8318 8318 0 0 0 0 0

Gyre Anticosti (0.5) 16630 14535 14534 1 2096 0 50 551

Gyre Anticosti (337) 8328 8328 8328 0 0 0 0 0

Havre-St-Pierre (0.5) 7865 7865 7865 0 0 0 0 0

Havre-St-Pierre (1) 7865 7865 7865 0 0 0 0 0

Havre-St-Pierre (120) 7865 7865 7865 0 0 0 0 0

IML (0.5) 8156 8156 8156 0 0 0 0 0

IML PEM (12.4) 15643 15643 15643 0 0 0 0 0

Ile Shag (1) 6238 6238 6238 0 0 0 0 0

Ile Shag (10) 5851 5851 5851 0 0 0 0 0

La Perle (1) 11111 11111 11111 0 0 0 0 0

La Romaine (1) 7920 7920 7920 0 0 0 0 0

La Romaine (2) 7920 7920 7920 0 0 0 0 0

La Romaine (21.9) 7920 7920 7920 0 0 0 0 0

La Tabatière (1) 6534 6534 6534 0 0 0 0 0

La Tabatière (39) 6534 6534 6534 0 0 0 0 0

Mont-Louis (0.5) 8404 8404 8404 0 0 0 0 0

Mont-Louis (30) 8404 8404 8404 0 0 0 0 0

Natashquan (1) 7921 7921 7921 0 0 0 0 0

Natashquan (5.4) 7057 7057 7057 0 0 0 0 0

Port-Menier (2) 8453 8453 8453 0 0 0 0 0

49

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Port-Menier (12.8) 8453 8453 8453 0 0 0 0 0

Rimouski (0.5) 18692 18690 18687 3 5 0 4 2

Rimouski (330) 9346 9346 9346 0 0 0 0 0

Riv.-au-Tonnerre (1) 7795 7795 7795 0 0 0 0 0

Riv.-au-Tonnerre (16) 7795 7795 7795 0 0 0 0 0

Sept-Iles (1) 8799 8799 8799 0 0 0 0 0

Sept-Iles (2) 8799 8799 8799 0 0 0 0 0

Sept-Iles (21.9) 8799 8799 8799 0 0 0 0 0

Shediac Valley (0.5) 16693 16693 16684 9 9 0 6 3

Shediac Valley (85.8) 8347 8347 8347 0 0 0 0 0

Tadoussac (1) 9407 9407 9407 0 0 0 0 0

Tadoussac (36.6) 9407 9407 9407 0 0 0 0 0

Table A2.17 Number of records and gaps in water temperature time series starting in 2009. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 8286 8286 8286 0 0 0 0 0

Baie-Comeau (82.3) 17766 17766 17766 0 0 0 0 0

Banc Beaugé (0.5) 14053 14041 14041 0 12 0 12 1

Banc Beaugé (97) 7026 7026 7026 0 0 0 0 0

Belle Isle (71) 11339 11339 11339 0 0 0 0 0

Bic (1) 8168 8168 8168 0 0 0 0 0

Bic (2) 8168 8168 8168 0 0 0 0 0

Bic (30.5) 8168 8168 8168 0 0 0 0 0

Blanc-Sablon (1) 5574 5574 5574 0 0 0 0 0

Blanc-Sablon (22) 5574 5574 5574 0 0 0 0 0

Borden (1.25) 16863 16863 16863 0 0 0 0 0

Courant de Gaspé (0.5) 16415 16451 16414 37 1 0 1 1

Courant de Gaspé (165) 8208 8208 8208 0 0 0 0 0

Grande-Rivière (2) 7539 7539 7539 0 0 0 0 0

Grande-Rivière (10) 7539 7539 7539 0 0 0 0 0

Gyre Anticosti (0.5) 16263 8984 8982 2 7281 0 1 7281

Gyre Anticosti (337) 8132 8132 8132 0 0 0 0 0

Havre-St-Pierre (0.5) 7148 7148 7148 0 0 0 0 0

Havre-St-Pierre (1) 7148 7148 7148 0 0 0 0 0

Havre-St-Pierre (120) 7148 7148 7148 0 0 0 0 0

IML (0.5) 8161 8161 8161 0 0 0 0 0

IML PEM (12.4) 6797 6797 6797 0 0 0 0 0

IML PEM (12.4) 35694 35694 35694 0 0 0 0 0

Ile Shag (1) 6207 6207 6207 0 0 0 0 0

Ile Shag (10) 5855 5855 5855 0 0 0 0 0

Ile Shag (10) 7807 7807 7807 0 0 0 0 0

Irving Whale (67) 17737 17737 17737 0 0 0 0 0

La Perle (1) 9948 9948 9948 0 0 0 0 0

La Perle (26) 9305 9305 9305 0 0 0 0 0

La Romaine (1) 7026 7026 7026 0 0 0 0 0

La Romaine (2) 7026 7026 7026 0 0 0 0 0

La Romaine (21.9) 7026 7026 7026 0 0 0 0 0

La Tabatière (1) 6904 6904 6904 0 0 0 0 0

La Tabatière (39) 6904 6904 6904 0 0 0 0 0

50

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Mont-Louis (0.35) 8396 8396 8396 0 0 0 0 0

Mont-Louis (0.5) 8396 8396 8396 0 0 0 0 0

Mont-Louis (1.1) 8396 8396 8396 0 0 0 0 0

Mont-Louis (2.1) 8396 8396 8396 0 0 0 0 0

Mont-Louis (30) 8396 8396 8396 0 0 0 0 0

Natashquan (1) 7208 7208 7208 0 0 0 0 0

Natashquan (5.4) 7208 7208 7208 0 0 0 0 0

Port-Menier (2) 8063 8063 8063 0 0 0 0 0

Port-Menier (12.8) 8063 8063 8063 0 0 0 0 0

Rimouski (0.5) 18354 18044 18044 0 310 0 146 80

Rimouski (330) 9177 9177 9177 0 0 0 0 0

Riv.-au-Tonnerre (1) 7169 7169 7169 0 0 0 0 0

Riv.-au-Tonnerre (16) 7169 7169 7169 0 0 0 0 0

Sept-Iles (2) 8354 8354 8354 0 0 0 0 0

Sept-Iles (21.9) 8354 8354 8354 0 0 0 0 0

Shediac Valley (0.5) 8094 7416 7416 0 678 0 99 82

Shediac Valley (85.8) 7514 7514 7514 0 0 0 0 0

Tadoussac (0.5) 9592 9592 9592 0 0 0 0 0

Tadoussac (36.6) 9592 9592 9592 0 0 0 0 0

Table A2.18 Number of records and gaps in water temperature time series starting in 2010. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 8974 8974 8974 0 0 0 0 0

Baie-Comeau (82.3) 17137 17137 17137 0 0 0 0 0

Banc Beaugé (0.5) 16539 16450 16450 0 89 0 89 1

Banc Beaugé (1) 8269 8269 8269 0 0 0 0 0

Banc Beaugé (97) 8269 8269 8269 0 0 0 0 0

Belle Isle (71) 17520 17520 17520 0 0 0 0 0

Bic (1) 8717 8717 8717 0 0 0 0 0

Bic (2) 8717 8717 8717 0 0 0 0 0

Bic (30.5) 8717 8717 8717 0 0 0 0 0

Blanc-Sablon (1) 8222 8222 8222 0 0 0 0 0

Blanc-Sablon (22) 8222 8222 8222 0 0 0 0 0

Borden (1.25) 6296 6296 6296 0 0 0 0 0

Borden (1) 32535 32535 32535 0 0 0 0 0

Courant de Gaspé (0.5) 14836 14819 14792 27 44 0 44 1

Courant de Gaspé (165) 9308 9308 9308 0 0 0 0 0

Grande-Rivière (2) 8964 8964 8964 0 0 0 0 0

Grande-Rivière (10) 8964 8964 8964 0 0 0 0 0

Gyre Anticosti (0.5) 18626 18713 18623 90 3 0 3 1

Gyre Anticosti (337) 9312 9312 9312 0 0 0 0 0

Havre-St-Pierre (0.5) 8461 8461 8461 0 0 0 0 0

Havre-St-Pierre (1) 8461 8461 8461 0 0 0 0 0

Havre-St-Pierre (120) 8461 8461 8461 0 0 0 0 0

IML (0.5) 8639 8639 8639 0 0 0 0 0

IML PEM (12.4) 15706 15706 15706 0 0 0 0 0

Ile Shag (1) 7345 7345 7345 0 0 0 0 0

Ile Shag (10) 7345 7345 7345 0 0 0 0 0

Ile Shag (10) 17758 17758 17758 0 0 0 0 0

51

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Irving Whale (0.5) 9886 9886 9886 0 0 0 0 0

Irving Whale (67) 9708 9708 9708 0 0 0 0 0

La Perle (1) 10393 10393 10393 0 0 0 0 0

La Perle (26) 10393 10393 10393 0 0 0 0 0

La Romaine (1) 8358 8358 8358 0 0 0 0 0

La Romaine (2) 8357 8357 8357 0 0 0 0 0

La Romaine (21.9) 8357 8357 8357 0 0 0 0 0

La Tabatière (1) 8253 8253 8253 0 0 0 0 0

La Tabatière (39) 8255 8255 8255 0 0 0 0 0

Mont-Louis (0.5) 9302 9302 9302 0 0 0 0 0

Mont-Louis (1.1) 9302 9302 9302 0 0 0 0 0

Mont-Louis (2.1) 9302 9302 9302 0 0 0 0 0

Mont-Louis (30) 9302 9302 9302 0 0 0 0 0

Mont-Louis (200) 50672 50672 50672 0 0 0 0 0

Natashquan (1) 8428 8428 8428 0 0 0 0 0

Natashquan (5.4) 8428 8428 8428 0 0 0 0 0

Port-Menier (2) 8783 8783 8783 0 0 0 0 0

Port-Menier (12.8) 8783 8783 8783 0 0 0 0 0

Rimouski (0.5) 18821 18748 18748 0 73 0 67 3

Rimouski (330) 9365 9365 9365 0 0 0 0 0

Riv.-au-Tonnerre (1) 8506 8506 8506 0 0 0 0 0

Riv.-au-Tonnerre (16) 8506 8506 8506 0 0 0 0 0

Sept-Iles (1) 8984 8984 8984 0 0 0 0 0

Sept-Iles (2) 8984 8984 8984 0 0 0 0 0

Sept-Iles (21.9) 8984 8984 8984 0 0 0 0 0

Shediac Valley (0.5) 17730 17734 17724 10 6 0 6 1

Shediac Valley (85.8) 8865 8865 8865 0 0 0 0 0

Tadoussac (0.5) 58211 58211 58211 0 0 0 0 0

Tadoussac (1) 9702 9702 9702 0 0 0 0 0

Tadoussac (36.6) 9702 9702 9702 0 0 0 0 0

Table A2.19 Number of records and gaps in water temperature time series starting in 2011. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 8769 8769 8769 0 0 0 0 0

Baie-Comeau (82.3) 17168 17168 17168 0 0 0 0 0

Banc Beaugé (0.5) 14318 14318 14318 0 0 0 0 0

Banc Beaugé (1) 14318 14318 14318 0 0 0 0 0

Belle Isle (71) 35221 35221 35221 0 0 0 0 0

Bic (1) 8440 8440 8440 0 0 0 0 0

Bic (2) 8440 8440 8440 0 0 0 0 0

Bic (30.5) 16881 16881 16881 0 0 0 0 0

Blanc-Sablon (1) 8778 8778 8778 0 0 0 0 0

Blanc-Sablon (22) 8778 8778 8778 0 0 0 0 0

Borden (1.25) 32391 32391 32391 0 0 0 0 0

Courant de Gaspé (0.5) 16683 16643 16629 14 54 0 54 1

Courant de Gaspé (165) 8341 8341 8341 0 0 0 0 0

Grande-Rivière (2) 8403 8403 8403 0 0 0 0 0

Grande-Rivière (10) 8402 8402 8402 0 0 0 0 0

Grande-Rivière-exp (1) 15692 15692 15692 0 0 0 0 0

52

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Grande-Rivière-exp (1) 15692 15692 15692 0 0 0 0 0

Grande-Rivière-exp (10) 48754 48754 48754 0 0 0 0 0

Grande-Rivière-exp (2) 15692 15692 15692 0 0 0 0 0

Gyre Anticosti (0.5) 16395 16395 16395 0 0 0 0 0

Gyre Anticosti (337) 8197 8197 8197 0 0 0 0 0

Gyre Anticosti-exp (10) 16395 16395 16395 0 0 0 0 0

Havre-St-Pierre (0.5) 15927 15927 15927 0 0 0 0 0

Havre-St-Pierre (1) 7964 7964 7964 0 0 0 0 0

Havre-St-Pierre (120) 7964 7964 7964 0 0 0 0 0

Havre-St-Pierre-exp (10) 15928 15928 15928 0 0 0 0 0

IML (0.5) 8359 8359 8359 0 0 0 0 0

IML PEM (12.4) 14766 14766 14766 0 0 0 0 0

Ile Shag (1) 15249 15249 15249 0 0 0 0 0

Ile Shag (10) 7625 7625 7625 0 0 0 0 0

Ile Shag (10) 20639 20639 20639 0 0 0 0 0

Irving Whale (0.5) 12889 12889 12889 0 0 0 0 0

Irving Whale (67) 25778 25778 25778 0 0 0 0 0

La Perle (1) 10695 10695 10695 0 0 0 0 0

La Romaine (1) 9054 9054 9054 0 0 0 0 0

La Romaine (2) 9054 9054 9054 0 0 0 0 0

La Romaine (21.9) 9054 9054 9054 0 0 0 0 0

La Tabatière (1) 8978 8978 8978 0 0 0 0 0

La Tabatière (39) 8978 8978 8978 0 0 0 0 0

Mont-Louis (0.35) 8205 8205 8205 0 0 0 0 0

Mont-Louis (0.5) 16409 16409 16409 0 0 0 0 0

Mont-Louis (1.1) 8205 8205 8205 0 0 0 0 0

Mont-Louis (2.1) 8205 8205 8205 0 0 0 0 0

Mont-Louis (30) 8205 8205 8205 0 0 0 0 0

Natashquan (1) 7880 7880 7880 0 0 0 0 0

Natashquan (5.4) 7880 7880 7880 0 0 0 0 0

Port-Menier (2) 8327 8327 8327 0 0 0 0 0

Port-Menier (12.8) 8327 8327 8327 0 0 0 0 0

Rimouski (0.5) 17952 17929 17929 0 23 0 23 1

Rimouski (330) 8976 8976 8976 0 0 0 0 0

Rimouski-exp (10) 17951 17951 17951 0 0 0 0 0

Riv.-au-Tonnerre (1) 8007 8007 8007 0 0 0 0 0

Riv.-au-Tonnerre (16) 8008 8008 8008 0 0 0 0 0

Sept-Iles (1) 8754 8754 8754 0 0 0 0 0

Sept-Iles (2) 8754 8754 8754 0 0 0 0 0

Sept-Iles (21.9) 8754 8754 8754 0 0 0 0 0

Tadoussac (0.5) 9454 9454 9454 0 0 0 0 0

Tadoussac (1) 9454 9454 9454 0 0 0 0 0

Tadoussac (36.6) 9454 9454 9454 0 0 0 0 0

Table A2.20 Number of records and gaps in water temperature time series starting in 2012. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 9572 9572 9572 0 0 0 0 0

Baie-Comeau (82.3) 18637 18637 18637 0 0 0 0 0

Banc Beaugé (0.5) 7768 7766 7766 0 2 0 2 1

53

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Banc Beaugé (1) 7768 7768 7768 0 0 0 0 0

Banc Beaugé (97) 7768 7768 7768 0 0 0 0 0

Belle Isle (71) 31594 31594 31594 0 0 0 0 0

Bic (1) 8673 8673 8673 0 0 0 0 0

Bic (2) 8673 8673 8673 0 0 0 0 0

Bic (30.5) 8673 8673 8673 0 0 0 0 0

Blanc-Sablon (1) 7852 7852 7852 0 0 0 0 0

Blanc-Sablon (22) 7851 7851 7851 0 0 0 0 0

Borden (1.25) 13377 13377 13377 0 0 0 0 0

Courant de Gaspé (0.5) 18583 18583 18583 0 0 0 0 0

Courant de Gaspé (165) 9289 9289 9289 0 0 0 0 0

Grande-Rivière (0.5) 9116 9116 9116 0 0 0 0 0

Grande-Rivière (2) 9116 9116 9116 0 0 0 0 0

Grande-Rivière (10) 9116 9116 9116 0 0 0 0 0

Havre-St-Pierre (0.5) 8117 8117 8117 0 0 0 0 0

Havre-St-Pierre (1) 8117 8117 8117 0 0 0 0 0

Havre-St-Pierre (120) 8116 8116 8116 0 0 0 0 0

Havre-St-Pierre-exp (10) 8117 8117 8117 0 0 0 0 0

Hudson-3 (159.4) 17988 17988 17988 0 0 0 0 0

IML (0.5) 8669 8669 8669 0 0 0 0 0

IML PEM (12.4) 17520 17520 17520 0 0 0 0 0

Ile Shag (1) 6912 6912 6912 0 0 0 0 0

Ile Shag (10) 6911 6911 6911 0 0 0 0 0

Ile Shag (10) 10474 10474 10474 0 0 0 0 0

Irving Whale (0.5) 12036 12036 12036 0 0 0 0 0

Irving Whale (67) 12036 12036 12036 0 0 0 0 0

La Perle (1) 10949 10949 10949 0 0 0 0 0

La Perle (26) 10942 10942 10942 0 0 0 0 0

La Romaine (1) 7792 7792 7792 0 0 0 0 0

La Romaine (2) 7792 7792 7792 0 0 0 0 0

La Romaine (21.9) 7792 7792 7792 0 0 0 0 0

La Tabatière (1) 7629 7629 7629 0 0 0 0 0

La Tabatière (39) 7629 7629 7629 0 0 0 0 0

Mont-Louis (0.35) 9268 9268 9268 0 0 0 0 0

Mont-Louis (0.5) 9268 9268 9268 0 0 0 0 0

Mont-Louis (1.1) 9268 9268 9268 0 0 0 0 0

Mont-Louis (2.1) 9268 9268 9268 0 0 0 0 0

Mont-Louis (30) 9268 9268 9268 0 0 0 0 0

Natashquan (1) 8085 8085 8085 0 0 0 0 0

Natashquan (5.4) 8085 8085 8085 0 0 0 0 0

Port-Menier (2) 9363 9363 9363 0 0 0 0 0

Port-Menier (12.8) 9363 9363 9363 0 0 0 0 0

Rimouski (0.5) 18763 18730 18729 0 34 1 11 17

Rimouski (330) 9381 9381 9381 0 0 0 0 0

Rimouski-exp (10) 9381 9381 9381 0 0 0 0 0

Riv.-au-Tonnerre (1) 8975 8975 8975 0 0 0 0 0

Riv.-au-Tonnerre (16) 8975 8975 8975 0 0 0 0 0

Sept-Iles (1) 9504 9504 9504 0 0 0 0 0

Sept-Iles (2) 9504 9504 9504 0 0 0 0 0

Sept-Iles (21.9) 9504 9504 9504 0 0 0 0 0

Shediac Valley (0.5) 8255 8255 8255 0 0 0 0 0

Shediac Valley (85.8) 8255 8255 8255 0 0 0 0 0

Tadoussac (0.5) 9349 9349 9136 0 213 213 1 213

Tadoussac (1) 9349 9349 9136 0 213 213 1 213

54

Table A2.21 Number of records and gaps in water temperature time series starting in 2013. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 8049 8049 8049 0 0 0 0 0

Baie-Comeau (82.3) 19303 19303 19303 0 0 0 0 0

Banc Beaugé (0.5) 7297 7290 7290 0 7 0 7 1

Banc Beaugé (1) 7297 7297 7297 0 0 0 0 0

Banc Beaugé (97) 7297 7297 7297 0 0 0 0 0

Belle Isle (71) 40922 40922 40922 0 0 0 0 0

Bic (1) 8450 8450 8450 0 0 0 0 0

Bic (2) 8450 8450 8450 0 0 0 0 0

Bic (30.5) 8449 8449 8449 0 0 0 0 0

Blanc-Sablon (1) 7398 7398 7398 0 0 0 0 0

Blanc-Sablon (22) 7398 7398 7398 0 0 0 0 0

Borden (1.25) 15954 15954 15954 0 0 0 0 0

Courant de Gaspé (0.5) 15426 15426 15426 0 0 0 0 0

Grande-Rivière (0.5) 7785 7785 7785 0 0 0 0 0

Grande-Rivière (2) 7785 7785 7785 0 0 0 0 0

Grande-Rivière (10) 7785 7785 7785 0 0 0 0 0

Havre-St-Pierre (0.5) 7539 7539 7539 0 0 0 0 0

Havre-St-Pierre (1) 7539 7539 7539 0 0 0 0 0

Havre-St-Pierre (120) 7539 7539 7539 0 0 0 0 0

Havre-St-Pierre-exp (10) 7539 7539 7539 0 0 0 0 0

Hudson-1 (268.4) 17559 17559 17559 0 0 0 0 0

Hudson-3 (159.4) 17844 17844 17844 0 0 0 0 0

IML (0.5) 8450 8450 8450 0 0 0 0 0

IML (14) 8450 8450 8450 0 0 0 0 0

IML PEM (12.4) 21565 21565 21565 0 0 0 0 0

Ile Shag (1) 7238 7238 7238 0 0 0 0 0

Ile Shag (10) 7238 7238 7238 0 0 0 0 0

Ile Shag (10) 11524 11524 11524 0 0 0 0 0

Irving Whale (0.5) 9509 9509 9509 0 0 0 0 0

Irving Whale (67) 9509 9509 9509 0 0 0 0 0

La Perle (1) 10227 10227 10227 0 0 0 0 0

La Perle (26) 10227 10227 10227 0 0 0 0 0

La Romaine (1) 5194 5194 5194 0 0 0 0 0

La Romaine (2) 5194 5194 5194 0 0 0 0 0

La Tabatière (1) 7233 7233 7233 0 0 0 0 0

La Tabatière (39) 7233 7233 7233 0 0 0 0 0

Mont-Louis (0.35) 8124 8124 8124 0 0 0 0 0

Mont-Louis (0.5) 8123 8123 8123 0 0 0 0 0

Mont-Louis (1.1) 8128 8128 8128 0 0 0 0 0

Mont-Louis (2.1) 8124 8124 8124 0 0 0 0 0

Mont-Louis (30) 8124 8124 8124 0 0 0 0 0

Natashquan (1) 7477 7477 7477 0 0 0 0 0

Natashquan (5.4) 7476 7476 7476 0 0 0 0 0

Port-Menier (2) 8062 8062 8062 0 0 0 0 0

Port-Menier (12.8) 8062 8062 8062 0 0 0 0 0

Rimouski (0.5) 16787 15860 15860 0 927 0 119 509

Rimouski (330) 8393 8393 8393 0 0 0 0 0

Rimouski-exp (10) 8393 8393 8382 0 11 11 1 11

Riv.-au-Tonnerre (1) 6889 6889 6889 0 0 0 0 0

55

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Riv.-au-Tonnerre (16) 6889 6889 6889 0 0 0 0 0

Sept-Iles (1) 8072 8072 8072 0 0 0 0 0

Sept-Iles (2) 8072 8072 8072 0 0 0 0 0

Sept-Iles (21.9) 8071 8071 8071 0 0 0 0 0

Shediac Valley (0.5) 7722 7722 7722 0 0 0 0 0

Shediac Valley (85.8) 7722 7722 7722 0 0 0 0 0

Tadoussac (0.5) 8978 8978 8978 0 0 0 0 0

Tadoussac (1) 8978 8978 8978 0 0 0 0 0

Tadoussac (36.6) 8978 8978 8978 0 0 0 0 0

Table A2.22 Number of records and gaps in water temperature time series starting in 2014. Each

line represents results from one raw file. Station name (depth, m) is followed by the expected

number of records, Ne, the number of valid records, Nv, the number of good records, Ng, the

number of oversampling records, No, the total number of missing records, Ntm, the number of

flagged records, Nf, the number of gap, NOG, and the maximum gap, MG.

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Baie-Comeau (1) 6502 6502 6502 0 0 0 0 0

Banc Beaugé (0.5) 11441 11426 11426 0 15 0 15 1

Banc Beaugé (1) 11439 11439 11439 0 0 0 0 0

Banc Beaugé (97) 5720 5720 5720 0 0 0 0 0

Bic (1) 7186 7186 7186 0 0 0 0 0

Bic (2) 7186 7186 7186 0 0 0 0 0

Bic (30.5) 7186 7186 7186 0 0 0 0 0

Blanc-Sablon (1) 5675 5675 5675 0 0 0 0 0

Blanc-Sablon (22) 5675 5675 5675 0 0 0 0 0

Borden (1.25) 17461 17461 17461 0 0 0 0 0

Courant de Gaspé (0.5) 13152 13096 13095 1 57 0 1 57

Courant de Gaspé (165) 6576 6576 6576 0 0 0 0 0

Grande-Rivière (0.5) 6569 6569 6569 0 0 0 0 0

Grande-Rivière (2) 6569 6569 6569 0 0 0 0 0

Grande-Rivière (10) 6569 6569 6569 0 0 0 0 0

Havre-St-Pierre (0.5) 5660 5660 5660 0 0 0 0 0

Havre-St-Pierre (1) 5660 5660 5660 0 0 0 0 0

Havre-St-Pierre (120) 5659 5659 5659 0 0 0 0 0

IML (0.5) 7208 7208 7208 0 0 0 0 0

IML (14) 7208 7208 7208 0 0 0 0 0

Ile Shag (1) 5714 5714 5714 0 0 0 0 0

Ile Shag (10) 5713 5713 5713 0 0 0 0 0

Irving Whale (0.5) 11388 11388 11388 0 0 0 0 0

Irving Whale (67) 11419 11419 11419 0 0 0 0 0

La Perle (1) 11181 11181 11181 0 0 0 0 0

La Romaine (1) 5845 5845 5845 0 0 0 0 0

La Romaine (2) 5845 5845 5845 0 0 0 0 0

La Romaine (21.9) 5846 5846 5846 0 0 0 0 0

La Tabatière (1) 5812 5812 5812 0 0 0 0 0

La Tabatière (39) 5812 5812 5812 0 0 0 0 0

Mont-Louis (0.35) 6575 6575 6575 0 0 0 0 0

Mont-Louis (0.5) 6576 6576 6574 0 2 2 1 2

Mont-Louis (1.1) 6576 6576 6576 0 0 0 0 0

Mont-Louis (30) 6575 6575 6575 0 0 0 0 0

Natashquan (1) 5914 5914 5914 0 0 0 0 0

Natashquan (5.4) 5914 5914 5914 0 0 0 0 0

Old Harry (0.5) 13481 10194 10191 0 3290 3 82 3208

56

Station (depth, m) Ne Nv Ng No Ntm Nf NOG MG Rimouski (0.5) 17381 16893 16875 1 506 17 35 269

Rimouski (330) 8788 8788 8788 0 0 0 0 0

Riv.-au-Tonnerre (1) 6042 6042 6042 0 0 0 0 0

Riv.-au-Tonnerre (16) 6042 6042 6042 0 0 0 0 0

Sept-Iles (1) 6479 6479 6479 0 0 0 0 0

Sept-Iles (2) 6479 6479 6479 0 0 0 0 0

Shediac Valley (0.5) 13142 13138 13138 0 4 0 1 4

Shediac Valley (85.8) 6571 6571 6571 0 0 0 0 0

Tadoussac (0.5) 45504 45504 45504 0 0 0 0 0

Tadoussac (1) 45506 45506 45506 0 0 0 0 0

Tadoussac (36.6) 7584 7584 7584 0 0 0 0 0

57

APPENDIX 3. CHARTS OF DATA AVAILABILITY

Figure A3.1 Data chart for 1993. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

Figure A3.2 Data chart for 1994. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

58

Figure A3.3 Data chart for 1995. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

.

59

Figure A3.4 Data chart for 1996. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

60

Figure A3.5 Data chart for 1997. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

61

Figure A3.6 Data chart for 1998. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

62

Figure A3.7 Data chart for 1999. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

63

Figure A3.8 Data chart for 2000. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

64

Figure A3.9 Data chart for 2001. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

65

Figure A3.10 Data chart for 2002. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

66

Figure A3.11 Data chart for 2003. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

67

Figure A3.12 Data chart for 2004. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

68

Figure A3.13 Data chart for 2005. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

69

Figure A3.14 Data chart for 2006. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

70

Figure A3.15 Data chart for 2007. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

71

Figure A3.16 Data chart for 2008. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

72

Figure A3.17 Data chart for 2009. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

73

Figure A3.18 Data chart for 2010. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

74

Figure A3.19 Data chart for 2011. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

75

Figure A3.20 Data chart for 2012. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

76

Figure A3.21 Data chart for 2013. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.

77

Figure A3.22 Data chart for 2014. The horizontal line below the station name (depth, m) shows

the valid range of records. The yearly record number No (oversampled, magenta plus sign), Ntm

(total missing, blue dot), Nf (flagged, green x) and MG (maximum gap, red asterisk) are

respectively indicated only when No or Ntm is non-zero. The right and left pointing triangles

respectively indicate the start and end of a time series.